Patterns of Honey–Mumford learning-style scores and classifications among pre-service teachers in Ghana

Understanding the learning preferences of pre-service teachers is important for designing flexible and inclusive teacher education programmes. This study examined Honey and Mumford Learning Styles Questionnaire (LSQ) dimension scores and the resulting highest-scoring and tied maximum-score classifications among Level 100 pre-service teachers in Colleges of Education in the Ashanti Region of Ghana. It also examined whether these classifications were associated with gender, teaching-level specialisation, and programme family. A cross-sectional survey design was adopted. The accessible population comprised approximately 2,800 pre-service teachers across eight Colleges of Education, from which 1,496 respondents were included in the study. Data were collected using the 80-item Honey and Mumford LSQ, which assesses four learning-preference dimensions: Activist, Reflector, Theorist, and Pragmatist. For descriptive classification, the dimension with the highest observed score was identified for each respondent, while respondents with equal maximum scores on two or more dimensions were retained in a separate Tie category. Data were analysed using descriptive statistics, Pearson correlations, chi-square tests, Cramér's V, and multinomial logistic regression. Tied maximum scores were observed for 44.5% of respondents. Among respondents with a unique highest-scoring dimension, Activist was the most frequent classification (28.7%), followed by Theorist (12.0%), Reflector (9.5%), and Pragmatist (5.3%). Highest-scoring classification was significantly associated with gender, χ 2 (4, N = 1,496) = 17.91, p = .001, Cramér's V = .109; teaching level, χ 2 (4, N = 1,496) = 41.76, p < .001, Cramér's V = .167; and programme family, χ 2 (12, N = 1,496) = 178.94, p < .001, Cramér's V = .200. The effect sizes were relatively small. Multinomial regression identified differences in selected outcome contrasts, particularly for teaching level and programme family, although several estimates had wide confidence intervals and should be interpreted cautiously. The findings demonstrate considerable variation in LSQ score patterns and show that a substantial proportion of respondents could not be assigned a unique highest-scoring dimension. These results support cautious use of learning-style classifications as descriptive indicators of reported preferences rather than as fixed learner types. Teacher education programmes should therefore emphasise varied pedagogical experiences incorporating active engagement, reflection, theoretical understanding, and practical application.