Translation of the mental health literacy questionnaire for young adults into Chichewa for use in Malawi: preliminary validation and reliability results

Abstract

Abstract Background Mental Health Literacy (MHL) is the ability to recognise mental disorders, have knowledge of professional help available, effective self-help strategies, skills to give support to others, and knowledge of how to prevent mental disorders. Sufficient MHL is linked to better help seeking behaviour and management of mental illness. Assessing MHL importantly helps identify knowledge gaps and inaccurate beliefs about mental health issues, whilst informing development and better evaluation of MHL interventions. This study aimed to translate the English version of a self-reporting Mental Health Literacy questionnaire (MHLq) for young adult populations (16–30 years-old) into Chichewa for use in Malawi and evaluate the psychometric properties of this Chichewa version. Methods An established translation methodology was employed, involving back-translation, comparison, forward-translation, comparison, and piloting. The translated Chichewa questionnaire was initially piloted with 14 young adults in a Malawi university, then subsequently administered to 132 young adults in rural community settings across Malawi. Results Overall internal consistency of the Chichewa translated MHLq was good (Cronbach’s alpha = 0.67) although subscales’ scores ranged from acceptable (factor 1 and 3) to unacceptable (factor 2 and 4). Confirmatory factor analysis found Factor 1 (Knowledge of mental health problems), Factor 3 (First aid skills and help seeking behaviour) and Factor 4 (Self-help strategies) of the Chichewa version fit very well with related factors of the original English MHLq. For Factor 2 (Erroneous beliefs/stereotypes), 5 out of its 8 items correlated well with the original version. This suggests a four-factor solution is a reasonably good fit to the data. Conclusions Use of the Malawian MHLq among Chichewa speaking young adult populations is well supported for factors 1 and 3 but not for factors 2 and 4. More psychometric testing with a larger sample is vital to further validate the questionnaire. Further research is needed to carry out test/re-test reliability statistics.