Validating an assessment and feedback instrument for use in dietetics education: construct validity of the mini clinical evaluation exercise (Mini-CEX).
Anna Kleppe Moe, Helene Dahl, Monika Kvernenes, Aslaug Drotningsvik, Hanne Rosendahl-Riise
Abstract
Open AccessBACKGROUND: Dietetics education lacks standardised tools for formative assessment and feedback in clinical placement for students, which impairs learning outcomes. The Mini Clinical Evaluation Exercise (Mini-CEX) is a seven-item instrument designed for clinical supervisors' direct observation and assessment of student performance on six clinical competencies plus overall competence. It has been validated in contexts of several disciplines, but not in dietetics education. In this study we aim to assess the construct validity of Mini-CEX for this context. METHODS: We developed three videotapes portraying dietetics students carrying out patient consultations in secondary health care. An assessment panel of dietitians reached consensus regarding the level of competence portrayed by each student. Clinical dietitians were then invited to watch the videos individually in randomised order and assess the students' clinical competencies using the Mini-CEX instrument. Construct validity was investigated by analysing agreement between assessment panel and participants' scores, interrater reliability, and rank correlation between competencies. RESULTS: Clinical dietitians (n = 65) working in secondary health care in Norway participated in the study. On a scale of 1-9, the assessment panel's scores on the three students' overall clinical competence were 2, 5 and 7, participants mean scores were 1.9, 4.9 and 7.6, respectively. The proportion of participants' scores agreeing with the assessment panel scores ranged from 63 to 97% on the six competencies and overall competence. Interrater reliability between participants' scores was substantial (Krippendorff's alpha = 0.772). There were significant positive rank correlations between the competency variables in all videos (Kendalls' tau = 0.49-0.86, p < 0.0001). CONCLUSIONS: Our results indicate that participants were able to discriminate between dietetics students' levels of clinical competence using the Mini-CEX instrument in agreement with an assessment panel. The observed agreement and interrater reliability of the Mini-CEX scores are in favour of its construct validity in dietetics education. A valid observation and assessment using Mini-CEX constitutes an important step of the formative assessment process. Consensus amongst assessors and training in how to provide feedback is needed to ensure successful implementation.