Peer-reviewed international research papers published open-access with EOI assignment and global indexing across engineering, computer science, environmental science, social sciences, and more.
Learning disabilities can significantly affect educational achievement, yet conventional identification may be time-consuming and dependent on specialist assessment. Machine learning (ML) offers an opportunity to analyse learner behavioural and performance data and provide early risk indicators. This study investigates the impact of ML on learning-disability detection using dyslexia as an empirical case. The analysis uses the uploaded Dyt-desktop dataset containing 3,644 observations, 196 predictors and 392 dyslexia-positive cases. Random Forest (RF), XGBoost and Extra Trees (ET) were evaluated using stratified five-fold cross-validation. Random-Forest Recursive Feature Elimination (RF-RFE) was subsequently applied, followed by probability-level ensemble learning. Results show that XGBoost was the strongest individual base model, achieving 90.64% accuracy, 64.41% precision, 29.08% recall, 98.06% specificity, 40.07% F1-score, 0.8845 ROC-AUC and 0.3912 MCC. RF-RFE + XGBoost improved performance to 90.81% accuracy, 31.89% recall and 0.8858 ROC-AUC. The ensemble model substantially increased recall to 71.17% and MCC to 0.4644, although accuracy declined to 85.70%. The findings demonstrate that ML can provide useful early-warning capabilities for learning-disability screening, but accuracy alone is inadequate under class imbalance. The study concludes that ML should complement rather than replace professional educational assessment.
Introduction : Le partographe est un outil de surveillance maternelle et fœtale pendant la phase active du travail d’accouchement et d’aide à la décision en cas d’anomalies. Objectif : Notre objectif était d’évaluer la tenue du partogramme à la maternité du centre de santé de référence de Koutiala. Méthodologie : Notre étude était transversale et analytique allant de janvier à décembre 2022. Résultats : Environ 80,3 % des partogrammes étaient remplis par les sage-femmes. Les paucipares représentaient 46,8 % de l’étude. Le partogramme était rempli dans 87 % des cas dès l’admission de la parturiente. Le rythme horaire d’examen était respecté sur 92% des partogrammes contre 8%. La souffrance fœtale aiguë était l’anomalie la plus fréquente au cours du travail, avec 5 % des cas. Une césarienne d’urgence a été réalisée dans 10,5 % et l’extraction instrumentale (ventouse) a été réalisé dans 3% des cas. Les partogrammes incomplètement rempli représentaient 81% des cas. Plus la durée de travail est inférieure à 12 h, nous avons davantage un score d’Apgar compris entre 8 et 10 à la première mn (p = 0,006). Le pronostic maternel était satisfaisant. Conclusion : L’amélioration de la qualité du remplissage du partographe est le gauge de l’amélioration du pronostic maternel et fœtal. La sensibilisation, la formation et le suivi des agents quant à l’intérêt du partographe peuvent contribuer à sa meilleure utilisation.
Mots clés : Evaluation, partogramme, Koutiala
The proverbs occupy a central place in the African companies with oral tradition, where they constitute at the same time vectors of wisdom, social regulatory instruments and mobilizable argumentative resources in a variety of communicative situations.This study examines the contextual applicability of the proverbs in language cisalampasu, spoken by the people salampasu about the territory about Luiza (province of Kasaï Central, Democratic République of Congo).Starting from a corpus of 50 proverbs collected near 15 old advisors from 62 to 87 years, the analysis mobilizes the theoretical executives of pragmatic (theory of the acts of language of Austin and Searle, theory of the relevance of Sperber and Wilson) and of linguistic anthropology (model SPEAKING of Hymes).The results reveal that the use of the proverbs cisalampasu is conditioned by a series of contextual factors:the statute of the interlocutors, the nature of the situation (palaver, education, mourning, rejoicing), stake of the communication (persuasion, dissuasion, council, judgement) and the cultural standards which govern the word in public.The analysis shows that the proverbs salampasu are not applicable fixed statements in all circumstances, but of the discursive resources on which the effectivity depends on a fine adequacy between the selected statement and the situation of stating.This study contributes to the documentation of a language hitherto little studied and to the comprehension of the pragmatic mechanisms to work in the African oral traditions.
Dyslexia is a specific learning disorder that affects reading, decoding and spelling and may result in persistent academic difficulties when not identified early. Machine learning (ML) provides an opportunity to support early identification by detecting predictive patterns in behavioural learning data. This study comparatively evaluates three ensemble-learning paradigms—bagging, boosting and stacking—for dyslexia prediction using the Dyt-desktop dataset. The dataset contains 3,644 observations, 196 predictors, 392 dyslexia-positive cases and 3,252 non-dyslexia cases. Random Forest (RF) and Extra Trees (ET) were evaluated as bagging approaches, XGBoost as a boosting approach, and a heterogeneous RF-XGBoost-ET model as a stacking approach. Random-Forest Recursive Feature Elimination (RF-RFE) was additionally used to investigate the effect of feature reduction. Models were evaluated using accuracy, precision, recall, specificity, F1-score, ROC-AUC and Matthews Correlation Coefficient (MCC). XGBoost achieved the strongest base-model performance, obtaining 90.64% accuracy, 29.08% recall, 40.07% F1-score, 98.06% specificity and 0.8845 ROC-AUC. RF-RFE + XGBoost improved performance to 90.81% accuracy, 31.89% recall, 42.74% F1-score, 97.91% specificity, 0.8858 ROC-AUC and 0.4122 MCC. The stacking ensemble produced the highest recall (71.17%), F1-score (51.71%) and MCC (0.4644), although its accuracy decreased to 85.70% and specificity to 87.45%. The findings demonstrate that bagging, boosting and stacking optimize different aspects of dyslexia prediction. Boosting provided the strongest overall individual-model performance, while stacking was substantially more effective for sensitivity-oriented screening. Consequently, model selection should be based on the intended screening objective rather than accuracy alone
Oluwasheye Paul Folorunsho; Babalola Julius Daodu; Deborah Adesoye Ogunlade; Remi-Esan Sumbo Sherifat; and Sulaiman-Oloko Abdul (Correspondence Author)
Abstract
This paper analyzes how climate physical hazards exposure drives performance of commercial bank credit portfolios in eight sub-Saharan African (SSA) economies of Kenya, Ethiopia, Uganda, Tanzania, Nigeria, Ghana, Côte d’Ivoire and Senegal for the period 2010-2023. A Climate Exposure Index (CEI) is created by overlaying geospatial hazard maps on banks geographic exposure of their lending portfolios, and embedded in a dynamic GMM panel system framework complemented with a spatial econometric approach to capture spatial linkages across districts. One standard deviation in composite CEI leads to a 1.84 percentage points worsening of the non-performing loan ratio (NPL), a 2.31 percentage point increase in loan loss provision intensity and expanding credit margins in climate vulnerable sectors. Droughts are shown to be the primary trigger of agricultural loan distress in East Africa and floods drive credit deterioration in West Africa. Unambiguously, sampled banks price climate risk below actual anticipated levels and under-provision its impact; their provisioning deficit from the optimal levels based on our climate-adjusted expected loss estimates range from 23% to 47%. Policy relevant findings are provided to banking supervision, climate-friendly capital requirements, the design of macro-prudential buffers, and the operationalization of physical climate risk in African credit risk management framework.
ABSTRACT
The issuance by the Democratic Republic of Congo of its first sovereign Eurobond of 500 million USD in December 2021, concurrently with the relaunch of the Kinshasa Stock Exchange, raises the critical question of the articulation between external financing and the development of the domestic capital market in frontier economies.
Drawing on a mixed-methods approach combining panel data from 28 Eurobond-issuing countries in Sub-Saharan Africa over the period 2010-2024, an event study, and a process-tracing case study of the DRC, this article demonstrates that the impact of a Eurobond on a nascent stock exchange is entirely conditional on the pre-existing institutional quality.
Fixed-effects econometric results indicate that a stock of Eurobonds exceeding 2% of GDP reduces domestic market capitalization by 0.18 points when the WGI governance index is below -0.58. The case study reveals three negative transmission channels in the DRC : liquidity absorption, the bypassing of democratic accountability, and the absence of a legal framework. By extending rentier state theory, we conceptualize "debt rentierism."
Beyond diagnosis, this article proposes, for the attention of Authorities, Institutions, Researchers, and Students, an "Institutional Safeguard Matrix of 7 Points" aimed at aligning future issuances with the development of the Kinshasa Stock Exchange and at avoiding the reproduction of the resource curse in the debt market.
Keywords : Sovereign Eurobond, Kinshasa Stock Exchange, Frontier Markets, Sovereign Debt, Governance, Debt Rentierism, DRC.
ABSTRACT
For 2,500 years, global history has been shaped by competition over strategic resources. The Democratic Republic of Congo, holding 70% of world cobalt reserves, has been trapped in this logic of predation. Using mixed-methods research with 2015-2025 panel data and 25 semi-structured interviews, this paper argues that conflicts in the DRC are contemporary expressions of economic warfare. Statistical results show a 0.87 correlation between security and mining FDI. Our regression model indicates that a 1-point increase in security leads to +182.74 million USD in FDI. We propose a 4-pillar "Laboratory of Economic Peace" model centered on securing 4 strategic trade corridors. The study concludes that peace in the DRC is the most profitable economic strategy for Central Africa.
Keywords : Geopolitics, DRC, Economic Peace, Trade Corridors, Critical Minerals, Security
Background: Fractures of the distal tibia outside the ankle joint present particular fixation challenges because the metaphyseal segment is broad and soft-tissue coverage is limited. Intramedullary nailing and minimally invasive locked plating are both established surgical options.
Objective: To compare functional recovery, fracture healing, alignment, and selected perioperative and postoperative outcomes after intramedullary nailing or distal tibial locked plating.
Patients and Methods: Thirty adults with closed extra-articular distal-third tibial fractures extending within two Müller squares were randomized equally to minimally invasive distal tibial locked plating or intramedullary nailing. Operative duration, C-arm image count, fibular stabilization, radiographic union, alignment, infection, reoperation, and Olerud-Molander Ankle Score were assessed. Statistical analysis used SPSS version 25, with P≤0.05 defining statistical significance.
Results: Surgery was completed more rapidly with intramedullary nailing than plating (70.00±13.62 vs 86.13±13.52 minutes; P=0.003), and fewer C-arm images were recorded (53.93±17.66 vs 66.53±12.39; P=0.033). Healing occurred at 18.93±7.41 weeks with nailing and 20.07±9.30 weeks with plating (P=0.725). Malalignment was seen in 2 (13.3%) nail cases versus none after plating (P=0.143). Two patients (13.3%) in each group developed infection (P=1.000). OMAS category distribution was comparable (P=0.425), as was the frequency of second surgery (13.3% vs 20.0%; P=0.169).
Conclusion: In this small randomized series, both constructs provided acceptable functional and radiographic results. Nailing was associated with shorter procedures and fewer recorded C-arm images, while no statistically significant between-group differences were found for union, OMAS, infection, alignment, or secondary surgery.
Peanut butter is a nutrient-dense product whose safety and acceptability can be influenced by the
source of groundnuts and by processing hygiene. This study evaluated the microbial quality and
sensory acceptability of peanut butter produced from three local groundnut market types: Solar,
Kampala, and White. The samples were sorted, roasted, blanched, milled, packaged separately, and
evaluated by standard culture-based methods. Total bacterial count (TBC), total coliform count
(TCC), and total fungal count (TFC) were determined, while representative isolates were
characterized using cultural, microscopic, Gram-stain, and biochemical tests. Sensory evaluation
was conducted with 10 panelists using a nine-point hedonic scale. TBC values were 7.3 × 10³, 8.2 ×
10³, and 6.5 × 10³ CFU/g for Solar, Kampala, and White samples, respectively. TFC values were
6.3 × 10³, 5.8 × 10³, and 2.9 × 10³ CFU/g, respectively. Coliforms were not detected in Solar or
Kampala samples but were enumerated at 1.9 × 10³ CFU/g in the White sample. Presumptive
isolates included Staphylococcus spp. and Bacillus cereus in Solar, Bacillus cereus in Kampala, and
Salmonella-like and Bacillus cereus isolates in White; Aspergillus-like fungi were observed in all
samples. Kampala had the highest overall mean sensory score (7.20 ± 1.43), closely followed by
Solar (7.22 ± 1.35), whereas White was least preferred (4.00 ± 2.13). The occurrence of coliforms
and presumptive food-safety-relevant organisms indicates the need for improved sanitation,
confirmatory identification, and mycotoxin testing before commercial distribution.
Keywords: peanut butter, groundnut, sensory evaluation, microbial quality, food safety, Aspergillus Spp.
Machine failures in manufacturing industries result in significant production losses, increased maintenance costs, and reduced equipment reliability. Predictive maintenance has emerged as an effective strategy for minimizing unplanned downtime by forecasting equipment failures before they occur. This study developed a machine learning-based predictive maintenance model for milling machine failure prediction using a Random Forest classifier. Operational data comprising air temperature, process temperature, rotational speed, torque, and tool wear were utilized as input variables. A dataset containing 10,000 observations was obtained and divided into training (70%) and testing (30%) subsets. The Random Forest model was trained using 100 decision tree estimators and evaluated using accuracy, precision, recall, F1-score, and confusion matrix metrics. The model achieved training accuracy, precision, recall, and F1-score values of 99.34%, 84.03%, 96.15%, and 89.69%, respectively. Testing results yielded accuracy, precision, recall, and F1-score values of 98.53%, 71.29%, 82.76%, and 76.60%, respectively. Confusion matrix analysis showed that the model correctly classified 2,884 non-failure cases and 72 failure cases. The results indicate that Random Forest is highly effective for milling machine failure prediction and can support proactive maintenance decisions, thereby reduce downtime and improve operational efficiency in manufacturing environments.
Protecting the environment is a shared responsibility. In the current global context, climate change and environmental protection have become central concerns. One of the ways the environment can be severely degraded is through mining operations.
This study aimed to assess community perspectives on the rehabilitation of abandoned mine sites. Data were collected through survey and questions focused on appropriate management strategies for such sites. The findings revealed that the community places significant trust in regulatory institutions and mining companies, holding them primarily responsible for managing abandoned mine sites.
Key aspects were explored to determine the most effective strategic approach to abandoned mine management: Governance, A large proportion of respondents expressed confidence in government institutions responsible for regulating and planning mining sector development. The community also advocated for transforming abandoned mines into sites for ecological restoration, ideally with economic co-benefits.
Funding for rehabilitation, there was a strong consensus that mining companies should bear the financial responsibility for rehabilitation efforts. Additionally, community members expressed a willingness to actively participate in initiatives aimed at restoring abandoned mine sites, reflecting a collaborative approach to environmental stewardship.
These insights underscore the importance of multi-stakeholder engagement combining regulatory oversight, corporate accountability, and community participation to achieve sustainable rehabilitation outcomes.
Key word: Abandoned Mine site rehabilitation, community perspective, Environmental rehabilitation, Multi-stakeholder engagement, Regulatory trust
This study tackles the social experience and psychological state of victims of juvenile delinquency in Kisangani. We observed that acts of violence committed by children and adolescents are increasing in this city. To conduct our research, we surveyed 123 victims. We used questionnaires, including the Cohen Perceived Stress Scale and the Hamilton State-Anxiety Scale. We also conducted semi-structured interviews. The results we obtained are quite worrying. Thus, 61 % of victims have a permanent feeling of insecurity. Furthermore, 57 % report a deterioration in their social ties. We also found that 46 % of respondents observe a weakening of their family structure. Psychologically, the figures are just as alarming: 70 % of victims have moderate to severe state-anxiety and 33 % have high perceived stress. Many victims suffer from hyperemotionality (67 %), irritability (62 %) or withdraw into themselves (55 %). Data analysis shows a positive correlation between negative social experience and psychological distress (r = 0.67, p < 0.01). These results confirm that acts of juvenile delinquency deeply affect the social and psychological lives of victims. The article suggests ways to improve victim support and strengthen prevention.
Keywords: Social experience ; psychological state ; juvenile delinquency ; victims ; Kisangani.
This research examines the dimensions of psychological state that predominate among victims of juvenile delinquency in Kisangani. It also analyzes their relationship with social experience. Faced with the increase in violent acts committed by children and adolescents in this city, we surveyed 123 victims. We used questionnaires (Cohen Perceived Stress Scale, Hamilton State-Anxiety Scale) and semi-structured interviews.Results indicate that the predominant psychological dimensions are state-anxiety (70% moderate to severe), perceived stress (33% high), and various symptoms such as hyperemotionality (67%), irritability (62%), and withdrawal (55%). Data analysis shows a positive correlation between these dimensions and negative social experience (r = 0.67, p < 0.01). The feeling of insecurity is the main predictor of psychological distress (β = 0.35). These results confirm that acts of juvenile delinquency deeply affect the psychological state of victims. The article suggests ways to improve psychological care for victims and strengthen support systems.
Keywords: Social experience ; psychological state ; juvenile delinquency ; victims ; Kisangani.
Learning disabilities can significantly affect educational achievement, yet conventional identification may be time-consuming and dependent on specialist assessment. Machine learning (ML) offers an opportunity to analyse learner behavioural and performance data and provide early risk indicators. This study investigates the impact of ML on learning-disability detection using dyslexia as an empirical case. The analysis uses the uploaded Dyt-desktop dataset containing 3,644 observations, 196 predictors and 392 dyslexia-positive cases. Random Forest (RF), XGBoost and Extra Trees (ET) were evaluated using stratified five-fold cross-validation. Random-Forest Recursive Feature Elimination (RF-RFE) was subsequently applied, followed by probability-level ensemble learning. Results show that XGBoost was the strongest individual base model, achieving 90.64% accuracy, 64.41% precision, 29.08% recall, 98.06% specificity, 40.07% F1-score, 0.8845 ROC-AUC and 0.3912 MCC. RF-RFE + XGBoost improved performance to 90.81% accuracy, 31.89% recall and 0.8858 ROC-AUC. The ensemble model substantially increased recall to 71.17% and MCC to 0.4644, although accuracy declined to 85.70%. The findings demonstrate that ML can provide useful early-warning capabilities for learning-disability screening, but accuracy alone is inadequate under class imbalance. The study concludes that ML should complement rather than replace professional educational assessment.
This article examines the role of prior communication and negotiation mechanisms in shaping interactions between the Congolese National Police and students at the University of Kisangani during student protests in 2019 and 2022. Rather than reducing confrontation to police use of force or protester behaviour alone, the study reconstructs the relational sequence preceding escalation: the existence of contact channels, timing of information, accessibility of interlocutors, reciprocity of exchanges, and the possibility of negotiated arrangements. The research adopts a qualitative comparative design combining documentary and media content analysis, thematic coding, and sequence reconstruction inspired by process tracing. The empirical corpus consists mainly of Radio Okapi and Agence congolaise de presse reports on the February 2019 and December 2022 episodes, supplemented by February and September 2022 sequences. Findings suggest that the central issue is not simply the absence of communication, but the limited functionality of prior mechanisms: late communication, weak reciprocity, poorly documented negotiation, or the inability of existing channels to produce arrangements before escalation. In both central cases, the most structured institutional responses become visible mainly after confrontation. The article concludes that communication insufficiency is a plausible contributing mechanism to the predominance of negative police–student interactions, while operating within a broader context of hostility, unresolved grievances and reciprocal contentious behaviour. The hypothesis is therefore provisionally supported and requires further triangulation with interviews and institutional records.
Keywords: police communication; negotiation; student protest; police–student interaction; public order; University of Kisangani; Democratic Republic of the Congo.
The Democratic Republic of the Congo (DRC) has been a central context for the application of contemporary international criminal justice. In 2004, Congolese authorities referred the situation to the International Criminal Court (ICC), marking an institutional response to widespread international crimes, particularly in Ituri and the eastern provinces. The period between 2010 and 2015 was especially significant, with major judicial developments including the conviction of Thomas Lubanga (2012), the acquittal of Mathieu Ngudjolo Chui (2015), the conviction of Germain Katanga (2014), and the opening of Bosco Ntaganda’s trial (2015).
This study examines the achievements and limitations of the ICC in addressing impunity for international crimes in the DRC during this period. Using a qualitative approach based on legal and documentary analysis of ICC decisions, United Nations reports, and academic literature, the research highlights both progress and constraints. Findings show that ICC intervention strengthened judicial accountability and contributed to jurisprudence on war crimes and crimes against humanity. However, its impact was limited by selective prosecutions, lengthy proceedings, difficulties in apprehending suspects, the weakness of the national justice system, and the disconnect between international justice and affected communities.
The study concludes that the ICC alone cannot sustainably address impunity in the DRC. Its effectiveness depends on genuine complementarity with an independent Congolese judiciary, adequately resourced and capable of prosecuting perpetrators of international crimes.
Keywords: International Criminal Court, Democratic Republic of the Congo, impunity, war crimes, crimes against humanity, complementarity.
Abstract
The prosecution of international crimes in the Democratic Republic of the Congo (DRC) occurs in a context marked by persistent armed conflicts, institutional fragility, and complex relations between political power and the judiciary. Between 2010 and 2015, the International Criminal Court (ICC) pursued several Congolese nationals, including Thomas Lubanga, Germain Katanga, and Bosco Ntaganda, while the acquittal of Mathieu Ngudjolo highlighted the rigor of international criminal proceedings. ICC intervention, however, underscores the tension between combating impunity and preserving Congolese sovereignty.
This study examines the political and security challenges shaping the prosecution of international crimes in the DRC during this period. Adopting a qualitative, legal, and documentary approach, it analyzes the Rome Statute, ICC jurisprudence, United Nations reports, and academic literature on complementarity and transitional justice. Findings indicate that ICC action strengthened accountability and reinforced the international norm against impunity. Yet, its effectiveness was constrained by reliance on State cooperation, selective prosecutions, weak national institutions, and security imperatives. The study concludes that sustainable justice in the DRC requires balanced interaction between sovereignty, international cooperation, and the strengthening of domestic judicial mechanisms.
Keywords: Democratic Republic of the Congo, International Criminal Court, impunity, sovereignty, security, complementarity.
This paper shows that the continuity of the term N(y) which is of the form f(x, y) , is not sufficient to solve the extra ordinary differential equation by Adomian decomposition method.
In this work we illustrate a property of 2k by 2 magic rectangle. The generalized inverse Moor/penrose (M/P) can be constucted simply, and this property can be seen.
The purpose of this note is to give a method for finding the nth roots of a three by three magic square matrix and four by four magic square matrix where n is odd number. Also this method is generalized to a three by three magix square matrix whose entries are also mtrices.
Several methods of computing sums of consecutive triangular numbers exist. This letter presents an easy method that uses Newton's difference formula for polynomials.
This study investigates the economic impacts of seasonal climate variability on agricultural productivity and rural livelihoods in Rwanda, especial Rwamagana district. Given the country’s reliance on rain-fed agriculture, fluctuations in seasonal weather patterns pose significant risks to crop yields, household incomes, and the broader economy. The primary objectives were to quantify the economic losses attributable to seasonal climate fluctuations and to assess their implications on rural livelihoods and national economic stability. Using a mixed-methods approach that combined quantitative analysis of household survey data with qualitative insights from key informants, the study reveals that irregular rainfall, prolonged dry spells, and temperature fluctuations have led to substantial economic losses in agriculture. The findings show that seasonal climate variability accounts for approximately 62% of the variation in agricultural productivity, resulting in an average loss of 15-20% in crop yields during high variability periods. These yield reductions translate into significant income declines for smallholder farmers, contributing to increased food insecurity and household poverty. The regression analysis further indicates that seasonal climate fluctuations directly reduce farm income by up to 18% and increase the cost of production by 12%, thereby undermining household profitability and overall economic resilience. Qualitative data corroborate these findings, highlighting farmers’ experiences of crop failures, reduced income, and increased vulnerability due to unpredictable seasonal patterns. The study emphasizes the urgent need for targeted adaptation strategies, including improved weather forecasting and climate-resilient farming practices, to mitigate these economic losses and safeguard Rwanda’s rural livelihoods and economic growth.
La transition entre l’université et le marché du travail constitue une période décisive pour les étudiants finissants, particulièrement dans les contextes où l’accès à l’emploi formel demeure incertain et où les opportunités d’insertion varient selon les domaines de formation. Cette étude examine les différences de perception de l’avenir professionnel, d’impuissance acquise et de sentiment d’auto-efficacité chez les étudiants finissants de l’Université de Kisangani, en fonction de leur faculté et de leur auditoire de formation. Une enquête quantitative transversale a été menée au moyen d’un questionnaire de 27 items administré aux étudiants de Master 2 et de quatrième année de doctorat en médecine. L’instrument comprend trois échelles relatives à la perception de l’avenir professionnel, à l’impuissance acquise et au sentiment d’efficacité personnelle. Les analyses reposent sur les statistiques descriptives.
Les résultats révèlent des différences statistiquement significatives selon l’auditoire et la faculté. Les étudiants de quatrième année de doctorat en médecine présentent une perception plus favorable de leur avenir professionnel (M=29.78), un niveau plus faible d’impuissance acquise (M=19.95) et un niveau plus élevé d’auto-efficacité (M=43.67) que les étudiants de Master 2. Les comparaisons entre facultés montrent également des écarts significatifs pour les trois dimensions étudiées. Les étudiants en médecine se distinguent par un profil psychosocial plus favorable, tandis que certaines filières, notamment l’économie, le droit, la psychologie et la gestion des ressources naturelles, présentent des indicateurs de vulnérabilité relativement plus élevés. Ces résultats plaident en faveur de dispositifs d’orientation, de préparation à l’emploi et d’accompagnement psychosocial différenciés selon les facultés.
La transition entre l’université et le marché du travail constitue une période de vulnérabilité psychologique pour les étudiants finissants, en particulier lorsque les opportunités d’emploi sont perçues comme limitées, incertaines ou difficilement contrôlables. Cette étude examine les liens entre l’impuissance acquise, la perception de l’avenir professionnel et le sentiment d’auto-efficacité chez les étudiants finissants de l’Université de Kisangani. Elle teste plus spécifiquement le rôle modérateur de l’auto-efficacité dans la relation entre l’impuissance acquise et la perception de l’avenir professionnel.
Une enquête quantitative transversale est menée grâce à un questionnaire de 27 items évaluant trois construits. Les sous-échelles présentent une cohérence interne satisfaisante. Les corrélations de Pearson révèlent une association négative entre l’impuissance acquise et la perception de l’avenir professionnel (r=−0,407), une association positive entre l’auto-efficacité et la perception de l’avenir (r=0,624), ainsi qu’une association négative entre l’auto-efficacité et l’impuissance acquise (r=−0,403). L’analyse de modération met en évidence un effet principal négatif de l’impuissance acquise (β=−0,176), un effet principal positif de l’auto-efficacité (β=0,547) et une interaction négative significative entre l’impuissance acquise et l’auto-efficacité (β=−0,227).
L’analyse des pentes simples montre que l’association négative entre impuissance et avenir professionnel est non significative à faible auto-efficacité, significative à un niveau moyen et plus marquée à forte auto-efficacité. Les résultats suggèrent que, dans un contexte de contraintes professionnelles perçues comme structurelles, l’auto-efficacité ne fonctionne pas systématiquement comme un facteur protecteur. Elle peut accroître la sensibilité psychologique des étudiants à l’impuissance lorsqu’un fort sentiment de compétence entre en contradiction avec l’expérience de barrières perçues comme incontrôlables.
La transition entre l’université et le marché du travail constitue une période décisive pour les étudiants finissants, particulièrement dans les contextes où l’accès à l’emploi formel demeure incertain et où les opportunités d’insertion varient selon les domaines de formation. Cette étude examine les différences de perception de l’avenir professionnel, d’impuissance acquise et de sentiment d’auto-efficacité chez les étudiants finissants de l’Université de Kisangani, en fonction de leur faculté et de leur auditoire de formation. Une enquête quantitative transversale a été menée au moyen d’un questionnaire de 27 items administré aux étudiants de Master 2 et de quatrième année de doctorat en médecine. L’instrument comprend trois échelles relatives à la perception de l’avenir professionnel, à l’impuissance acquise et au sentiment d’efficacité personnelle. Les analyses reposent sur les statistiques descriptives.
Les résultats révèlent des différences statistiquement significatives selon l’auditoire et la faculté. Les étudiants de quatrième année de doctorat en médecine présentent une perception plus favorable de leur avenir professionnel (M=29.78), un niveau plus faible d’impuissance acquise (M=19.95) et un niveau plus élevé d’auto-efficacité (M=43.67) que les étudiants de Master 2. Les comparaisons entre facultés montrent également des écarts significatifs pour les trois dimensions étudiées. Les étudiants en médecine se distinguent par un profil psychosocial plus favorable, tandis que certaines filières, notamment l’économie, le droit, la psychologie et la gestion des ressources naturelles, présentent des indicateurs de vulnérabilité relativement plus élevés. Ces résultats plaident en faveur de dispositifs d’orientation, de préparation à l’emploi et d’accompagnement psychosocial différenciés selon les facultés.
La transition entre l’université et le marché du travail constitue une période de vulnérabilité psychologique pour les étudiants finissants, en particulier lorsque les opportunités d’emploi sont perçues comme limitées, incertaines ou difficilement contrôlables. Cette étude examine les liens entre l’impuissance acquise, la perception de l’avenir professionnel et le sentiment d’auto-efficacité chez les étudiants finissants de l’Université de Kisangani. Elle teste plus spécifiquement le rôle modérateur de l’auto-efficacité dans la relation entre l’impuissance acquise et la perception de l’avenir professionnel.
Une enquête quantitative transversale est menée grâce à un questionnaire de 27 items évaluant trois construits. Les sous-échelles présentent une cohérence interne satisfaisante. Les corrélations de Pearson révèlent une association négative entre l’impuissance acquise et la perception de l’avenir professionnel (r=−0,407), une association positive entre l’auto-efficacité et la perception de l’avenir (r=0,624), ainsi qu’une association négative entre l’auto-efficacité et l’impuissance acquise (r=−0,403). L’analyse de modération met en évidence un effet principal négatif de l’impuissance acquise (β=−0,176), un effet principal positif de l’auto-efficacité (β=0,547) et une interaction négative significative entre l’impuissance acquise et l’auto-efficacité (β=−0,227).
L’analyse des pentes simples montre que l’association négative entre impuissance et avenir professionnel est non significative à faible auto-efficacité, significative à un niveau moyen et plus marquée à forte auto-efficacité. Les résultats suggèrent que, dans un contexte de contraintes professionnelles perçues comme structurelles, l’auto-efficacité ne fonctionne pas systématiquement comme un facteur protecteur. Elle peut accroître la sensibilité psychologique des étudiants à l’impuissance lorsqu’un fort sentiment de compétence entre en contradiction avec l’expérience de barrières perçues comme incontrôlables.
Cet article met en lumière la contradiction de ressources énergétiques de la RD Congo, un pays doté d’un potentiel hydroélectrique exceptionnel estimé à 100 000 MW, mais dont une grande partie de la population demeure privée d’électricité. À travers une approche fondée sur la théorie de l’État prédateur, nous soutenons que cette situation ne résulte pas uniquement de contraintes techniques, mais d’une gouvernance défaillante caractérisée par la politisation et la mauvaise gestion de la Société Nationale d’Électricité (SNEL). En nous s’appuyant sur des données couvrant la période 2014-2024, notre étude montre que malgré une capacité hydroélectrique installée, la production effective reste très faible et les pertes sur le réseau demeurent élevées. Cette inefficacité se traduit par un faible taux d’accès à l’électricité, notamment dans les zones rurales, accentuant les inégalités territoriales et sociales.
L’article démontre également que l’insuffisance de l’électricité engendre un « multiplicateur social de la précarité », c’est-à-dire un ensemble d’effets négatifs qui se liés ensembles dans les domaines de la santé, de l’éducation, de l’emploi et de l’environnement.
Organizational commitment, work addiction and burnout are major constructs in occupational psychology whose interactions considerably influence employees' psychological health and organizational performance. Although these variables have been extensively investigated in industrialized countries, empirical evidence remains limited in the context of the Democratic Republic of Congo. This study aimed to examine the relationships between organizational commitment, work addiction and burnout among employees working in public and private enterprises in the city of Goma.
A quantitative, descriptive, correlational and cross-sectional research design was adopted. The study involved 364 employees selected from public and private organizations. Data were collected using standardized psychometric instruments assessing organizational commitment, work addiction and burnout. Statistical analyses included descriptive statistics, Pearson correlation coefficients.
The findings reveal generally high levels of organizational commitment, work addiction and burnout among participants. Significant associations were observed between several dimensions of organizational commitment, work addiction and burnout. Affective and normative commitment were associated with lower levels of several work addiction dimensions, whereas emotional exhaustion emerged as the burnout component most strongly related to addictive work behaviors. Continuance commitment showed more complex and mixed relationships with the studied variables.
These findings emphasize the importance of fostering healthy organizational commitment as a strategy to improve employees' psychological well-being and prevent psychosocial risks. Public and private organizations should implement human resource management policies aimed at promoting supportive work environments, healthy employee engagement and burnout prevention.
Keywords: Organizational commitment; Work addiction; Workaholism; Burnout; Occupational psychology.
This article comparatively examines web journalism practices used by Kis24 and Kaba Lisolo to address the infodemic during an Ebola virus disease outbreak in Tshopo Province. The study is based on a corpus of ten items, equally distributed between the two media outlets. Framing theory is used to identify dominant communication functions, source categories and verification practices associated with health-news production. Findings show a clear predominance of the mobilization function, representing 47.6% of all coded framing occurrences, followed by factual framing (33.3%) and anti-rumour framing (19.0%). No didactic or advocacy framing was identified in the supplied data. Provincial authorities and physicians/experts are the most frequently cited source categories, with four occurrences each, followed by WHO with three. Both media outlets are reported as using source cross-checking, expert consultation, image/video verification, rumour verification, post-publication correction and formal fact-checking protocols. These findings suggest that local web journalism’s contribution to infodemic management is mainly grounded in public mobilization, reliance on institutional and expert sources, and verification mechanisms. However, the absence of didactic framing indicates a limited pedagogical dimension in the observed corpus.
Keywords: web journalism; infodemic; Ebola; framing; fact-checking; health communication; Kis24; Kaba Lisolo; Tshopo.
Face aux crises structurelles et aux mutations globales du XXIe siècle, le cadres traditionnel d’analyse macroéconomique s’avère insuffisant. Cet article examine la trajectoire historique et conceptuelle des politiques macroéconomiques, en théorisant le passage du célèbre « carré magique » de Nicholas Kaldor vers un « polygone stratégique des possibles ». Fondée sur la théorie de l’économie du bien-être et la correction des défaillances de marché, la politique économique contemporaine ne peut plus se restreindre aux seuls objectifs de croissance du PIB, de plein emploi, de stabilité des prix et d’équilibre extérieur. Elle doit désormais intégrer des impératifs multidimensionnels et interconnectés, notamment la transition écologique, la santé publique, la souveraineté numérique et la résilience. En combinant une approche institutionnelle à une formalisation géométrique inspirée de la topologie mathématique et des programmes d’optimisation primal-dual, ce travail modélise la déformation de l’espace stratégique sous l’effet des chocs conjoncturels. L’article opérationnalise ce modèle par le biais d’indicateurs de développement standardisés, transformant l’octogone obtenu en un véritable tableau de bord de l’efficience publique. Ce cadre renouvelé offre un instrument central de gouvernance systémique pour piloter des trajectoires de développement durables, inclusives et résilientes face à l’incertitude globale.
No papers found matching your search. Try a different keyword.
Published papers are indexed in
Submissions Open
See Your Research in the Next Edition
Join thousands of researchers published in GSJ. Submit your manuscript — peer review in 1–3 days, open access publication, EOI assignment, and global indexing across 150+ countries.
