Answers are grounded in this paper and cite their sources.
County governments in Kenya depend on information systems for public service delivery and invest in anti-malware technologies to protect sensitive information but effectiveness of these technologies is rarely evaluated. The second of three articles from the author’s MBA research evaluated anti-malware security technologies deployed by Homa Bay County Government. Effectiveness was operationalised as detection, prevention, incident frequency, response time, update and patching practices, user awareness and perceived effectiveness. Convergent mixed methods case study design was used. Data were collected in April–May 2025 through a structured questionnaire (150 distributed; 136 returned; 90.7%), semi-structured interviews with four ICT key informants and technical audit of 12 systems (2 servers; 10 workstations). Questionnaire data were analysed descriptively using descriptive statistics and multiple linear regression, interviews thematically and audit data descriptively. Anti-malware software was installed on all 12 audited systems and reported by 85.3% of respondents (Windows Defender for Endpoint (62.1%) and Kaspersky Endpoint Security (41.4%) were most often mentioned by those with software installed). Real time protection was enabled on all audited systems; automatic updates and cloud threat intelligence were enabled on 83.3%; heuristic detection was enabled on 75%; scheduled scans on 66%; operating systems updated fully on 83.3% and third party applications on 66%. Internal logs showed a detection rate of 95% compared to 88% in independent scans and average response time was two hours. Three of 42 detections (7.1%) were zero-day threats (without a controlled zero-day test this suggests but does not prove limited detection of unknown malware); perceptions were more positive than audit evidence. Effectiveness was rated moderate or high by 82.4% of respondents (57.4% satisfied) but perceptions were higher than evidence from the audit. The regression model explained 61.2% of variance in perceived effectiveness (R² = 0.612; F(6, 129) = 33.94; p < 0.001); staff training (β = 0.356) and patch management (β = 0.312) were the strongest predictors followed by user awareness, resource allocation and technology type and policy enforcement was not significant. The technologies provide a baseline protection against known threats but are moderately effective because configuration gaps exist, patching is inconsistent, products are fragmented and little evidence that emerging threats are detected. A phased programme is recommended: first correct configuration and patching within existing licences and budgets then standardise and centralise platform and add behaviour based detection and server protection as funds allow.
No takes yet. Share an insight, caveat, or question.
Okura Odhiambo James (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: