An Evaluation of Inter-coder and Intra-coder Reliability for 24-Hour Dietary Recall Data Entered in WebNEERS.

OBJECTIVE: To evaluate inter-coder (between-coder) and intra-coder (within-coder) reliability among trained data coders who enter 24-hour dietary recall data collected through Expanded Food and Nutrition Education Program operations in the state of Georgia. DESIGN: This study employed multiple cross-sectional evaluations of inter-coder reliability and a short-term longitudinal evaluation of intra-coder reliability. PARTICIPANTS/SETTING: Study participants consisted of trained data coders (n = 9) who were employed during the 12-month period of evaluation. MAIN OUTCOME MEASURES: Primary outcome measures were inter-coder and intra-coder reliability across data entered into the Web-based Nutrition Education Evaluation and Reporting System. Statistical analyses were conducted using IBM SPSS 24. Descriptive statistics were generated and inter-coder and intra-coder reliability were assessed using 2-way mixed intraclass correlation coefficients. RESULTS: Results of this evaluation indicated good to excellent inter-coder reliability among all coders, and excellent intra-coder reliability among the majority of coders. However, some notable inconsistencies were identified within the intra-coder reliability analyses. CONCLUSIONS: Future strategies to improve data quality within Expanded Food and Nutrition Education Program operations might include enhanced training for data coders, implementation of error detection protocols, expansion of the Web-based Nutrition Education Evaluation and Reporting System database, and exploration of automated, computer-assisted administration of 24-hour dietary recalls.