Training and Assessing Model for the Ability of Community Health Volunteers in Anthropometric Measurement Using the Rasch Stacking and Racking Analyses

Abstract

Background. Inappropriate anthropometric measurements of infants and toddlers lead to a misclassification in nutritional status and loss of important interventions. Considering the practice conducted in this program within a country, its impact on millions of children must be considered. This study assesses the ability of community health volunteers (CHVs) before and after anthropometric training on infants and toddlers.  Methods. This study used a quantitative approach with a quasiexperimental and pretest-posttest design. The pre- and posttraining assessments of CHVs were conducted by standardized trainers using instruments developed according to WHO standards. There were 11 and 13 statement items for infants’ and toddlers’ indicators of assessment in anthropometric measurements, respectively. The result of the assessment was then analyzed using Rasch modeling with stacking and racking data analysis techniques. Results. The CHVs’ skills before training were far from adequate. Although widely varied, all trainees improved their abilities. Stacking analysis showed that the skills of all CHVs in measuring infants and toddlers increased by 2.68 and 3.34 logits ( $p<0.01$ ), respectively. Racking analysis showed a decrease in the perceived difficulty of all items by 2.61 and 3.07 logits for infant and toddler measurements, respectively ( $p<0.01$ ). The results of the racking analysis showed that the difficulty in measuring the anthropometrics of infants decreased more than that of toddlers. Conclusions. CHVs’ capacity to monitor child growth must be refreshed regularly. Standardized and proper training and assessment were developed to make CHVs reliable in taking anthropometric measurements of infants and toddlers.