Preservice Teachers' Artificial Intelligence Competence: A Network and Latent Profile Analyses

ABSTRACT The rapid integration of artificial intelligence (AI) in education has increased the need for teachers who possess multidimensional AI competencies. However, empirical evidence on preservice teachers' AI competence remains limited, particularly within developing‐country contexts such as Sub‐Saharan Africa. This study examined preservice teachers' AI competence in Ghana by exploring the structural relationships among AI competence indicators, assessing overall competence levels, identifying latent competence profiles and determining whether gender differences exist across six domains of AI competence: ‘AI Knowledge, AI Pedagogy, AI Assessment, AI Ethics, Human‐Centred Education, and Professional Engagement’. A descriptive cross‐sectional survey design was adopted, and data were collected from 509 preservice teachers who had completed AI‐related modules in a Ghanaian university. The instrument was validated through confirmatory factor analysis, and data were analysed using network analysis (NA), descriptive statistics, latent profile analysis (LPA) and Mann–Whitney U tests. The network analysis revealed a moderately interconnected competence structure, with pedagogical and assessment‐related indicators functioning as central bridging nodes. Descriptive results indicated that preservice teachers reported generally high AI competence across all domains, with Human‐Centred Education and Professional Engagement recording the highest mean scores. The LPA identified two competence profiles: a smaller high‐competence group (38.5%) and a larger moderately high competence group (61.5%). Finally, the Mann–Whitney U tests revealed no statistically significant gender differences across the six competence domains. The findings highlight the multidimensional and heterogeneous nature of AI competence and underscore the need for differentiated training approaches to strengthen AI integration within teacher education programmes.