The research, conducted by Abokor Hassan of the University of Burao’s School of Postgraduate Studies and Research, surveyed 278 faculty members and administrative staff across four public higher education institutions: the University of Burao, the University of Hargeisa, Burao Central University College and Sanag University. These universities anchor Somaliland’s public higher education system, a system that operates under persistent strain. Reports from the country’s Ministry of Education and National Higher Education Commission describe chronic resource constraints, weak internal governance, limited strategic management capacity and inadequate funding, conditions that make employee commitment both harder to earn and more critical to sustain.

The study’s central question is deceptively simple: does transactional leadership make university employees more committed to their institutions, and if so, how? Organizational commitment, in the psychological sense, is not mere attendance or compliance. It reflects an employee’s emotional attachment to the organization, identification with its goals and values, and willingness to exert effort on its behalf. Decades of research link high commitment to better job performance, lower turnover intentions and stronger institutional effectiveness, which matters enormously for universities that depend on dedicated staff for teaching, research and student services.

To answer the question, the study drew on four complementary theoretical pillars. Social exchange theory, developed by Peter Blau, holds that employees reciprocate favorable treatment with positive attitudes and loyalty. Frederick Herzberg’s two-factor theory distinguishes intrinsic motivators such as achievement, recognition and responsibility from extrinsic hygiene factors such as salary, job security and working conditions. Cameron and Quinn’s competing values framework conceptualizes organizational culture along clan, adhocracy, market and hierarchy dimensions. Finally, Meyer and Allen’s three-component model classifies commitment into affective, continuance and normative forms, which the study combined into a single higher-order measure of overall commitment.

Methodologically, the research followed a quantitative, cross-sectional design analyzed with partial least squares structural equation modeling, or PLS-SEM, a statistical framework well suited to testing networks of relationships among latent psychological constructs. The four universities were purposively selected for their strategic importance and regional diversity, and a stratified random sampling procedure, sized with the adjusted Cochran formula for finite populations, distributed 320 questionnaires. After screening, 278 valid responses remained, an effective response rate of 86.9 percent. Transactional leadership was measured with items adapted from the Multifactor Leadership Questionnaire of Bass and Avolio, commitment with the three-component model questionnaire, culture with the Organizational Culture Assessment Instrument, and motivation with indicators grounded in Herzberg’s theory.

The statistical results tell a layered story. Transactional leadership showed a direct, positive and statistically significant effect on organizational commitment, with a path coefficient of 0.260. But its influence on the two proposed mediators was even stronger: the style explained 21.3 percent of the variance in organizational culture and 15.4 percent of the variance in employee motivation, with effect sizes in the medium range. In turn, both culture and motivation independently predicted commitment. Bootstrapping with 5,000 resamples confirmed significant indirect effects through each mediator, accounting for roughly 23.8 percent of the total effect through culture and 24.0 percent through motivation. Together, the two pathways carried nearly half of the leadership’s total impact, and because the direct effect remained significant, the study classified both as partial mediators.

The model’s explanatory and predictive credentials were tested with unusual rigor for this research area. The three predictors jointly explained 45.3 percent of the variance in organizational commitment, a moderate level, and blindfolding procedures confirmed predictive relevance for every endogenous construct. A PLS-predict analysis comparing the model against a linear regression benchmark found lower prediction errors for 19 of 20 indicators, indicating high out-of-sample predictive power. The researchers also confronted the classic weakness of single-source survey data, common method bias, with both procedural safeguards and statistical checks: Harman’s single-factor test showed the first factor explaining only 21.5 percent of variance, well below the 50 percent warning threshold, and all inner variance inflation factors fell under the conservative cutoff of 3. A Gaussian copula endogeneity assessment found no significant copula terms, suggesting the estimated paths are not distorted by reverse causality or omitted variables.

What do these numbers mean in practice? The picture that emerges is of transactional leadership as an architectural force rather than a merely transactional one. By clarifying role expectations, enforcing consistent performance feedback and administering transparent contingent rewards, leaders shape the shared norms and behavioral standards that constitute organizational culture. A cohesive culture, in turn, gives employees a sense of belonging and identification with institutional goals. Simultaneously, the same exchange mechanisms feed motivation: predictable rewards and fair recognition satisfy both the hygiene factors that prevent dissatisfaction and, through recognition and achievement, some of the intrinsic motivators that drive engagement. The finding that motivation carried a slightly larger share of the mediated effect than culture suggests that in resource-constrained settings, the psychological experience of being treated fairly and rewarded predictably may be the more immediate lever.

The study is candid about its limits. The cross-sectional design cannot establish causal ordering, and self-reported data from a single time point remain vulnerable to response bias despite the remedies applied. The sample was 73.7 percent male and skewed toward younger professionals, and the focus on public universities in one developing country limits generalizability to private institutions and other contexts. The transactional leadership scale also showed relatively modest internal consistency, with a Cronbach’s alpha of 0.680, prompting a call for further instrument validation and for qualitative approaches such as interviews and focus groups. Longitudinal and time-lagged designs are recommended to strengthen causal inference.

Even so, the implications reach beyond Somaliland. For university administrators, the study recommends institutionalizing transactional practices deliberately: clear performance contracts, transparent reward criteria and regular feedback mechanisms, paired with cultural initiatives such as shared-values workshops, participatory decision-making forums and open communication systems. Motivation strategies including recognition schemes, fair promotion criteria and non-monetary appreciation such as professional development opportunities round out the prescription. For policymakers, the findings argue for leadership development programs, performance management frameworks and incentive structures designed around the insight that structured, exchange-based leadership, far from being a motivational dead end, can generate durable psychological commitment when it is embedded in a supportive culture. In institutions where every retained lecturer and administrator counts, that is a finding worth taking seriously.

Subject of Research: The mediating roles of organizational culture and employee motivation in the link between transactional leadership and organizational commitment in Somaliland public universities

Article Title: The mediating role of organizational culture and employee motivation in the relationship between transactional leadership and organizational commitment

Article References: Hassan, A. (2026). The mediating role of organizational culture and employee motivation in the relationship between transactional leadership and organizational commitment. Discover Education, 5(1), Article 878. https://doi.org/10.1007/s44217-026-02108-2

Image Credits: AI Generated

DOI: 10.1007/s44217-026-02108-2

Keywords: transactional leadership, organizational commitment, organizational culture, employee motivation, PLS-SEM, higher education, Somaliland, public universities, social exchange theory, Herzberg two-factor theory, mediation analysis, organizational psychology

Breast cancer has become the most diagnosed malignancy in the world, with 2.3 million new cases and 670,000 deaths recorded in 2022, and projections suggest annual incidence could climb to 3.2 million cases by 2050. In sub-Saharan Africa, the burden is particularly punishing: late-stage presentation is common in Rwanda, Nigeria, Uganda, and Ethiopia, where limited treatment options and reliance on traditional medicine often delay diagnosis until the disease is advanced. Somaliland is no exception. The first cancer burden report from Hargeisa’s Needle Hospital identified breast cancer as the most common malignancy in the region, accounting for 15.9 percent of cases, with the majority of patients presenting at stage III or IV when survival odds are worst.

In settings where mammography machines are scarce and clinical breast examination infrastructure is thin, breast self-examination has long been championed as a cost-effective, non-invasive tool that empowers women to detect early pathological changes themselves. The technique is operationally defined as a three-step process: visual inspection of the breasts, palpation while lying down, and palpation while bathing, performed ideally three to five days after menstruation begins. The research team, led by Nafisa Abdihamid Mohamed of Amoud University, framed their investigation around the Health Belief Model, a psychological framework holding that people adopt preventive behaviors when they perceive a meaningful health threat and believe a specific action will effectively reduce it. Notably, the World Health Organization does not currently endorse self-examination as a population-based screening strategy, citing the absence of demonstrated mortality benefit and the risk of false positives and unnecessary biopsies, but the method remains influential in awareness campaigns across low-resource settings.

To quantify the drivers of practice in Borama, the researchers administered a structured, self-administered questionnaire between April and May 2024, using convenience sampling with proportional quotas to ensure representation from both institutions. The instrument was piloted among health science students, achieved a content validity index of 0.86 and a Cronbach’s alpha of 0.850, indicating solid validity and internal consistency. Of 217 students approached, 202 provided complete data, a response rate of 93.1 percent. The sample skewed young, with 82.7 percent of respondents aged 20 to 24, and predominantly single at 90.6 percent. Institutional quotas reflected enrollment figures, with 66.3 percent from Amoud University and 33.7 percent from Adal University, and participation was heaviest among third- and fourth-year students, who collectively made up more than half the cohort.

The descriptive results painted a picture of an exceptionally well-informed population. Fully 97 percent of respondents had received prior information about breast cancer or self-examination, drawn mainly from university instruction at 54.5 percent and social media at 24.3 percent. Some 63.4 percent correctly identified the recommended starting age of 20 to 25 years, 98.5 percent agreed that early detection improves survival, and 70.8 percent could name all the major risk factors, including family history, aging, unhealthy lifestyle, and hormonal therapy. Attitudinal measures were equally strong: every single respondent agreed that regular self-examination is important for women’s health, 99.5 percent believed the practice gives women control over their health, and 80.7 percent strongly agreed they felt confident encouraging other women to perform it.

Yet behavior lagged far behind belief. Only 94 of the 202 students, or 46.5 percent, had ever performed the examination. The 108 who had not cited the absence of symptoms rather than lack of awareness. Among those who did practice, timing adherence was encouraging, with 87.2 percent correctly examining themselves a few days after menstruation, but frequency was weaker, and intriguingly, 30.8 percent reported weekly examinations, exceeding the recommended monthly cadence, a pattern the authors suggest may reflect either proactive vigilance or misunderstanding of guidelines that warrants further investigation. Motivations for practice also revealed anxiety-driven triggers: 7.9 percent had previous breast problems, and 8.9 percent began the practice out of fear of breast cancer because of a family history.

The inferential analysis, conducted with binary logistic regression in SPSS while controlling for multicollinearity using variance inflation factors with a mean of 1.64, isolated the factors that truly predicted behavior. Age emerged as a powerful determinant: students in older age categories were more than five times as likely to perform self-examination as their 20-to-24-year-old peers, with an odds ratio of 5.012 and a confidence interval of 1.978 to 12.705. The authors attribute this to academic maturity and cumulative clinical exposure, which may heighten perceived severity as students witness late-stage cancer cases during their rotations. Knowledge, measured as a continuous score, contributed incrementally, with each additional point raising the odds of practice by 16.1 percent. Family monthly income, by contrast, showed no statistically significant association whatsoever, an odds ratio of 1.011 that diverges sharply from general population studies where economic status often dictates screening access.

Attitude proved to be the strongest predictor of all. Students with positive attitudes were 3.25 times more likely to perform the examination than those with poor or neutral attitudes, an odds ratio of 3.251 with a confidence interval of 2.214 to 4.774 and a p-value below 0.001. The finding carries a crucial caveat about psychological barriers: 18.3 percent of respondents reported feeling uncomfortable during the examination, and the authors interpret this discomfort, rooted in cultural modesty or fear of finding a lump, as a perceived barrier that can outweigh knowledge of personal susceptibility. In Health Belief Model terms, knowing you are at risk is not enough if the emotional cost of the action feels prohibitive. The authors caution that their framework was exploratory, since they measured only perceived benefits and barriers rather than deploying fully validated psychometric scales covering susceptibility, severity, and cues to action.

The study’s limitations are acknowledged candidly. The cross-sectional design precludes causal inference, self-reported practice was not clinically verified, and social desirability bias may have inflated prevalence figures among students aware of professional expectations. Convenience sampling from two urban universities also limits generalizability to remote and non-academic populations. Still, the pattern mirrors regional data from Ethiopia and Nigeria, and comparisons with a Mogadishu knowledge-attitude-practice study, in which 89 percent of healthcare professionals knew about breast cancer but awareness of preventive tools was strikingly limited, suggest the gap is systemic rather than local. The policy prescription is clear: Somaliland’s medical faculties should institutionalize hands-on simulations using breast models, mandate training for first- and second-year students before clinical pressures mount, and build peer-led advocacy programs that normalize the conversation around breast health, while the Ministry of Health reframes self-examination as an act of empowerment rather than a response to disease.

The deepest lesson of the Borama study may resonate far beyond Somaliland. Knowledge dissemination through lectures, posters, and social media campaigns is merely a precursor to behavior, not behavior itself. When 77.7 percent of a cohort knows and 46.5 percent act, the bottleneck is psychological, not informational. Bridging that divide demands interventions that build self-efficacy, desensitize discomfort through clinical practice, and transform abstract awareness into a sustained monthly habit. For the future physicians of Somaliland, who will be expected to lead their communities in early detection, the difference between knowing and doing could ultimately shape survival outcomes across an entire region where late-stage diagnosis remains the deadliest obstacle.

Subject of Research: Determinants of breast self-examination practice among female medical students in Somaliland

Article Title: Factors associated with breast self-examination practice among female medical students in selected universities in Borama, Somaliland

Article References: Factors associated with breast self-examination practice among female medical students in selected universities in Borama, Somaliland. (n.d.). https://doi.org/10.3389/fcacs.2026.1850142

Image Credits: AI Generated

DOI: 10.3389/fcacs.2026.1850142

Keywords: breast cancer, breast self-examination, Somaliland, medical students, Health Belief Model, early detection, cancer screening, knowledge-practice gap, cross-sectional study, health behavior, Amoud University, low-resource settings

The team deployed a mixed-methods, cross-sectional design that went well beyond a simple questionnaire. Household surveys were combined with key informant interviews, direct field observations, long-term climate records stretching back more than three decades, and secondary data on cereal production. Food insecurity was measured using the Household Food Insecurity Access Scale, or HFIAS, a validated instrument that grades households along a continuum from secure access to severe deprivation. Quantitative survey results were analyzed with descriptive statistics and bivariate association tests, interview material was subjected to thematic analysis, and climate records were examined using non-parametric trend analysis capable of detecting gradual shifts in temperature and rainfall behavior.

The headline numbers are stark. Only 19.3 percent of surveyed households were classified as food secure. A striking 74.6 percent were moderately food insecure, meaning they regularly compromise on the quality or quantity of the food they consume, while 6.1 percent were severely food insecure, experiencing hunger at a level that threatens health and productivity. Taken together, more than four in five households in the sample reported some degree of food insecurity, a prevalence that places Arabsiyo among the harder-pressed corners of the Horn of Africa’s drylands, where recurrent drought and volatile food prices have long strained subsistence economies.

When the researchers probed which household characteristics tracked with food security status, two factors emerged as statistically significant: market accessibility and marital status. Households with better access to markets were more likely to be food secure, a finding that underscores the importance of physical and economic connectivity in dryland farming systems, where the ability to sell surplus animals or purchase cereals during lean seasons can make the difference between stability and hunger. Marital status, meanwhile, appears to shape household composition, labor availability, and resource pooling, all of which influence a family’s capacity to withstand shocks. The association with education proved sensitive to the statistical test applied, a cautionary reminder that in small samples the choice of analytical method can determine which relationships appear robust.

Perhaps the most striking dimension of the study is the near-unanimity of climate perception among residents. Fully 89.5 percent of households reported noticing changes in temperature patterns, and 90.4 percent said droughts and floods had become more frequent. Some 85.1 percent observed shifts in the timing of planting and harvesting periods, 86.0 percent reported declining crop yields and reduced availability of staple foods, and 86.8 percent noted falling livestock productivity. In a community whose calendar, income, and diet are calibrated to the rhythm of the rains, such consensus is not anecdotal color; it is a collective diagnosis delivered by the people who live closest to the evidence.

Crucially, the instrumental record backs much of that testimony. Analyzing climate data from 1991 to 2023, the researchers found a statistically significant warming trend, quantified with a Mann-Kendall test that produced a tau statistic of 0.356 and a p-value of 0.004, indicating that the upward trajectory in temperature is very unlikely to be random fluctuation. Rainfall told a more complicated story: annual and seasonal totals displayed substantial variability from year to year, but no significant monotonic trend emerged. In other words, the rains have not simply disappeared; they have become erratic, arriving late, departing early, or failing in ways that disrupt planting schedules and pasture regeneration even when seasonal totals look superficially normal.

The qualitative strand of the research filled in the texture behind those statistics. Key informant interviews and field observations documented rainfall irregularity, chronic water scarcity, limited pasture availability, and persistent constraints on both crop and livestock production. For agro-pastoralists, who deliberately balance cultivation with herding to spread risk, such compound pressures are particularly corrosive. When rains fail, crops wither and pasture thins simultaneously, leaving families with few buffers. Livestock sales, the classic fallback, become less lucrative precisely when demand for slaughter animals falls and herds are weakened by poor grazing, creating a feedback loop that deepens food insecurity across the community.

The Arabsiyo findings resonate with a broader body of evidence from the Horn of Africa. Previous research in neighboring Ethiopia has documented similar patterns of smallholder vulnerability, coping strategies, and adaptation limits under climate variability, while studies across Somaliland have traced declining crop and livestock productivity in regions such as Marodijeh and Gabiley. Regional food security monitoring bodies have repeatedly flagged acute food insecurity in Somalia and its surroundings, and recent work has emphasized that socio-economic determinants, not climate alone, often dominate household food insecurity outcomes. The new study adds granular, district-level evidence to this picture, showing how global climate signals translate into household hunger through the mediating filters of market access, household structure, and livelihood design.

What distinguishes the research methodologically is its insistence on triangulation. Perceptions alone can be contested; statistics alone can be mute about lived experience. By aligning what residents say they observe with what thermometers and rain gauges actually recorded, and by cross-checking both against production data and on-the-ground observation, the authors built a case in which each line of evidence reinforces the others. The significant warming trend validates community reports of changing temperatures, while the absence of a rainfall trend, despite widespread perception of change, highlights how increased variability and unpredictability, rather than a simple decline in totals, may be the operative hazard for farming decisions.

The policy implications are correspondingly concrete. The authors point to improved market access as a priority, since connectivity emerged as one of the few statistically significant correlates of food security in the sample. They also emphasize the need for locally appropriate, climate-resilient agricultural and livelihood strategies, from water harvesting and rangeland management to crop and livestock interventions suited to semi-arid conditions. In a territory that remains internationally unrecognized and chronically under-resourced, the study is a reminder that adaptation planning must begin from the granular realities of specific districts. For the households of Arabsiyo, the data describe a present in which hunger is the norm and the climate is measurably warming; the question the research poses to planners and policymakers is whether interventions can arrive quickly enough to change that arithmetic.

Subject of Research: Household food insecurity and climate change perceptions among agro-pastoral communities in rural Arabsiyo District, Somaliland

Article Title: Household socio-demographics, climate change perceptions, and food security in agro-pastoral Somaliland: evidence from rural Arabsiyo District

Article References: Omer, M. A., & Mawlid, M. E. (2026). Household socio-demographics, climate change perceptions, and food security in agro-pastoral Somaliland: evidence from rural Arabsiyo District. SN Social Sciences, 6(10), Article 523. https://doi.org/10.1007/s43545-026-01826-6

Image Credits: AI Generated

DOI: 10.1007/s43545-026-01826-6

Keywords: food security, climate change, agro-pastoralism, Somaliland, Arabsiyo District, HFIAS, drought, rainfall variability, warming trend, market access, smallholder farmers, climate adaptation

The numbers behind that paradox are sobering. Sixty-five percent of students admitted to frequently reusing passwords or relying on weak ones, a habit that security experts consider one of the most dangerous practices in the digital world, since a single breach at one service can cascade across every account that shares the same credential. Nearly half of respondents, 48 percent, reported not using two-factor authentication on any of their online accounts, leaving a single stolen password as the only barrier between an attacker and their digital identity. Thirty percent said they would potentially click on links in unsolicited emails, the classic entry point for phishing attacks, and 40 percent demonstrated limited understanding of phishing and social engineering concepts altogether.

The study, published in Discover Education by Shuaib Jama Hassan of Amoud University and Abdisalan Ahmed Osman of Kabridahar University, went beyond simply cataloguing these risky behaviors. The researchers designed a cross-sectional survey with four sections covering demographics, a 20-question cybersecurity knowledge quiz, self-reported digital habits, and perceptions of institutional support. After cleaning the data, they combined theoretical knowledge and practical behavior into a single Weighted Awareness Score, weighting each component equally, and classified students into low, moderate, and high awareness categories. The results showed 38 percent of participants in the low awareness group, 34 percent in the moderate group, and just 28 percent demonstrating genuinely high awareness.

What makes this study methodologically interesting is its use of five different machine learning algorithms to predict awareness levels: Logistic Regression, Decision Tree, Random Forest, K-Nearest Neighbors, and Naive Bayes. The researchers deliberately chose these classical approaches over deep learning, arguing that with a dataset of just over a thousand tabular survey responses, interpretability and computational efficiency matter more than the raw power of neural networks. Each model was trained on 70 percent of the data and tested on the remaining 30 percent, with performance measured using accuracy, precision, recall, F1-score, and the area under the receiver operating characteristic curve.

The Random Forest model, an ensemble method that combines hundreds of decision trees to resist overfitting, emerged as the clear winner. It achieved an F1-score of 0.83 and an AUC-ROC of 0.92, indicating a strong balance between precision and sensitivity in identifying at-risk students. Logistic Regression followed with an AUC of 0.88, and K-Nearest Neighbors reached 0.86. The confusion matrix for the winning model showed most predictions concentrated along the diagonal, meaning the algorithm was particularly effective at distinguishing students with low and moderate awareness from their better-protected peers.

Perhaps the most actionable finding came from the feature importance analysis. The Random Forest model revealed that a student’s cybersecurity knowledge score carried the greatest predictive weight at 0.24, followed closely by prior ICT training at 0.21 and password reuse frequency at 0.18. Institutional support contributed 0.15, two-factor authentication adoption 0.12, and faculty type 0.10. Notably, demographic factors like age and gender carried far less weight than behavioral and educational variables, suggesting that vulnerability to cyber threats is not a matter of who students are but of what they have been taught and how they actually behave online.

The statistical analysis reinforced these findings. Students with prior ICT training were three times more likely to fall into the high awareness category, a difference so pronounced that the chi-square test yielded a value of 45.2 with a p-value below 0.001. ICT students scored an average of 68.5 on the knowledge quiz compared to 51.3 for their non-ICT peers, and the correlation analysis revealed a significant negative relationship between perceived institutional challenges and knowledge scores, meaning students who felt their universities failed to support digital safety tended to know less about it. In the logistic regression model, prior ICT training reduced the odds of inadequate awareness by 72 percent, while non-ICT enrollment combined with poor password habits tripled the likelihood of landing in the low awareness group.

By synthesizing the model’s most powerful features, the researchers identified three distinct risk profiles that read almost like a diagnostic manual for university administrators. Non-ICT students with no prior training and poor password habits formed the highest risk group, with 82 percent predicted to have low awareness. ICT students who had received training but maintained poor practical habits fell into the moderate risk group, with 45 percent predicted as low awareness. At the other end of the spectrum, ICT students with both training and good digital practices were predicted to have high awareness 92 percent of the time. These profiles offer a roadmap for institutions with limited resources, allowing them to concentrate interventions where they will have the greatest impact.

The researchers interpret the knowledge-behavior gap through the lens of Protection Motivation Theory, which holds that people act on threats only when they perceive them as personally relevant and feel capable of responding effectively. Somaliland’s students may intellectually recognize that cybercrime exists while underestimating their own likelihood of being targeted, or they may feel ill-equipped to implement security measures like two-factor authentication. This motivational gap, the authors argue, explains why awareness campaigns that simply transfer knowledge so often fail to change behavior. The theory suggests interventions must build both perceived threat relevance and practical self-efficacy, particularly among non-ICT students who showed the highest vulnerability in the study.

The study’s recommendations span two tiers of urgency. In the short term, universities can launch phishing simulation workshops with immediate feedback, hands-on password hygiene sessions, simplified two-factor authentication tutorials, peer mentoring programs in which ICT students coach their non-ICT classmates, and mobile-friendly educational resources suited to how students actually consume information. Longer-term structural changes include mandatory cybersecurity modules integrated across all faculties, institutional password policies, required two-factor authentication for sensitive university systems, and dedicated help desks where students can ask security questions without judgment. The authors also call for national coordination, urging Somaliland’s Ministry of Education and Science to establish minimum cybersecurity standards for tertiary institutions and a national digital literacy framework. The researchers acknowledge their limitations, including the cross-sectional design that prevents causal conclusions and the reliance on self-reported behavior, which is subject to social desirability bias. Still, as Somaliland’s universities rapidly digitize, this first-of-its-kind study provides both a warning and a practical playbook: cybersecurity resilience depends not just on what students know, but on whether their institutions make the safe choice the easy choice.

Subject of Research: Predicting cybersecurity awareness and risky online behavior among university students in Somaliland using machine learning models

Article Title: Predicting cybersecurity knowledge and behavior among Somaliland university students using machine learning

Article References: Hassan, S. J., & Osman, A. A. (2026). Predicting cybersecurity knowledge and behavior among Somaliland university students using machine learning. Discover Education, 5(1), Article 1142. https://doi.org/10.1007/s44217-026-02261-8

Image Credits: AI Generated

DOI: 10.1007/s44217-026-02261-8

Keywords: cybersecurity awareness, machine learning, Somaliland, higher education, phishing, password reuse, two-factor authentication, Random Forest, digital behavior, ICT training, risk profiles, Protection Motivation Theory

The research team, led by Nouh Botan and colleagues, conducted a retrospective cross-sectional review of routine microbiology records collected between January 1 and December 31, 2025. From the laboratory registry, they identified 286 positive bacterial culture records with antimicrobial susceptibility results, which together generated 1,787 individual test records because each isolate was challenged against multiple drugs. Applying the M39-A4 deduplication rule, which counts only the first unique isolate per patient per species to avoid statistical distortion from repeated sampling, the team excluded nine repeat isolates and arrived at a final analytic cohort of 277 unique clinical isolates. Intermediate susceptibility results were conservatively categorized as resistant for antibiogram purposes, in line with the guideline’s recommendations, while routine patient care continued to follow the laboratory’s standard reporting procedures.

The identity of the organisms tells its own story. Escherichia coli dominated the dataset, accounting for 50.9 percent of all isolates, followed by Staphylococcus aureus at 33.6 percent. Pseudomonas species made up 6.9 percent, and other enteric organisms, primarily Proteus and Klebsiella species, contributed 8.6 percent. Half of the isolates came from urine specimens and 39 percent from wound or pus samples, with blood cultures representing only 2.2 percent. The cohort was almost perfectly balanced by sex, and a slight majority of isolates originated from outpatient services. Notably, patient age and detailed clinical diagnoses were unavailable, because the retrospective laboratory database captured only microbiological surveillance variables, a limitation that reflects the sparse data infrastructure typical of the setting.

The susceptibility findings are where the study becomes genuinely alarming. Ampicillin, a legacy frontline agent, showed a susceptibility rate of just 7.3 percent among E. coli isolates, meaning it is essentially useless as an empirical option. Ciprofloxacin, a fluoroquinolone once considered a reliable workhorse, maintained susceptibility in only 19.0 percent of E. coli isolates, translating into an overall resistance burden of 81 percent. Among the broader Enterobacteriaceae pool, ciprofloxacin susceptibility fell to 16.3 percent. These figures align with disturbing trends reported from tertiary hospitals in Ethiopia, Kenya, and elsewhere in East Africa, where Gram-negative organisms have steadily accumulated resistance to fluoroquinolones, aminoglycosides, and third-generation cephalosporins.

Not every result was bleak. Nitrofurantoin, an older drug largely reserved for urinary infections, retained a susceptibility footprint of 78.5 percent among E. coli isolates and performed even better against S. aureus at 88.3 percent, supporting its continued role in empirical treatment of uncomplicated urinary tract infections. Gentamicin offered moderate coverage, with E. coli showing 60.4 percent susceptibility, although the Enterobacteriaceae pool fared worse at 51.1 percent. These patterns suggest that carefully chosen older, inexpensive agents can still deliver clinical value in the region, provided that prescribing decisions are anchored to locally generated susceptibility data rather than habit or availability.

Perhaps the most consequential finding concerns meropenem, a carbapenem antibiotic typically held in reserve as a last-line option for serious Gram-negative infections. Only 30.5 percent of tested E. coli isolates and 30.5 percent of the Enterobacteriaceae pool were classified as susceptible, with an overall susceptibility of 40.0 percent across organisms. The authors are careful to caution that reduced phenotypic susceptibility to meropenem should not be interpreted as evidence of specific carbapenemase-producing organisms without species-level identification and molecular confirmation, tools that were not available in this setting. Nevertheless, the signal is worrying enough that the researchers call urgently for confirmatory molecular characterization and continued prospective surveillance before any definitive epidemiological conclusions about carbapenem resistance in Somaliland can be drawn.

An exploratory subgroup analysis compared urinary and wound E. coli isolates and revealed intriguing heterogeneity. Wound isolates showed higher ciprofloxacin susceptibility than urinary isolates, 30.8 percent versus 17.2 percent, and higher gentamicin susceptibility at 68.2 percent versus 58.7 percent, but markedly lower meropenem susceptibility at 13.3 percent versus 35.4 percent. Because the subgroup sample sizes were small, the team used Fisher’s exact test and reported p-values of 0.267, 0.479, and 0.128 respectively, none reaching statistical significance. The authors emphasize that these comparisons are hypothesis-generating rather than conclusive, but they illustrate how specimen source can shape resistance profiles even within a single institution, a nuance that cumulative antibiograms stratified by specimen type can capture and communicate to prescribers.

Methodological honesty is a defining feature of the paper. Antibiotic-specific denominators varied throughout the dataset because susceptibility testing reflected routine clinical practice rather than a uniform research protocol: organism-specific panels, selective testing such as nitrofurantoin applied mainly to urinary isolates, individual clinician requests, and intermittent shortages of susceptibility disks and other consumables all meant that not every isolate was tested against every drug. Species-level identification relied on routine phenotypic and basic biochemical methods, since advanced techniques such as MALDI-TOF mass spectrometry and molecular assays were unavailable, and some organisms were therefore reported only at the genus level. Standardized daily internal quality control with ATCC reference strains could not be performed due to resource limitations, and breakpoint interpretation followed the CLSI- or EUCAST-based criteria accompanying whichever products were procurable. The authors explicitly acknowledge these constraints, arguing that transparent documentation of real-world laboratory conditions is itself a contribution toward building sustainable diagnostic capacity.

Beyond the numbers, the study is a proof of concept. The relatively modest culture volume of 286 positive records over a full year likely reflects broader diagnostic barriers: out-of-pocket costs for patients, heavy reliance on empirical therapy, intermittent supply constraints, and limited culture-guided prescribing. In Somaliland and neighboring Somali territories, healthcare is financed largely through direct patient payments, and antimicrobials are frequently obtainable without prescription oversight, while regional literature has documented incomplete counseling on dosing intervals and treatment duration. The authors also point to wider drivers of resistance that hospital data alone cannot capture, including unregulated retail antibiotic sales, veterinary antimicrobial use in pastoral livestock systems, and environmental exposures, arguing that long-term containment will require integrated collaboration across human health, veterinary medicine, and public health surveillance, consistent with a One Health approach.

What emerges from Hargeisa is both a warning and a template. The warning is that resistance to commonly used and even last-resort antibiotics is already substantial in a population previously invisible to global surveillance, and that without intervention, prescribers in the region are effectively practicing blind. The template is the demonstration that WHONET software, CLSI M39-A4 methodology, and disciplined deduplication of routine laboratory records can produce a credible, facility-specific antibiogram without expensive infrastructure. The researchers conclude that strengthening sustainable laboratory capacity, including reliable procurement of culture media, susceptibility testing materials, and equipment, should accompany routine antibiogram surveillance to support antimicrobial stewardship and evidence-based prescribing. For a region where no standardized cumulative antibiogram existed before, this baseline dataset is a small dataset with outsized significance, offering clinicians their first locally grounded map of which drugs still work and which no longer do.

Subject of Research: Antimicrobial resistance surveillance and cumulative antibiogram development at a tertiary referral hospital in Somaliland

Article Title: Standardized antimicrobial resistance surveillance in Somaliland: A hospital-based baseline evaluation using CLSI M39 and WHONET

Article References: Botan, N., Ali, A. S. O., Warsame, H. A., & Ali, Z. (2026). Standardized antimicrobial resistance surveillance in Somaliland: A hospital-based baseline evaluation using CLSI M39 and WHONET. New Microbes and New Infections, 74, Article 101861. https://doi.org/10.1016/j.nmni.2026.101861

Image Credits: AI Generated

DOI: 10.1016/j.nmni.2026.101861

Keywords: antimicrobial resistance, Somaliland, cumulative antibiogram, CLSI M39-A4, WHONET, Escherichia coli, meropenem, ciprofloxacin, nitrofurantoin, surveillance, Hargeisa, East Africa