IMPACT OF DATA CAPTURING METHODOLOGY ON DATA QUALITY: FOCUSING ON OPTICAL CHARACTER RECOGNITION (OCR) CASE OF OPTION B+ EVALUATION IN MALAWI

Abstract

This research investigated the impact of data collection methodology on data quality in the clinical research setting of Malawi's Prevention of Mother to Child Transmission Program (PMTCT). Employing questionnaires, interviews, and literature reviews, the study focused on Optical Character Recognition (OCR) technology for efficient data capture. Despite the methodology's benefits, including ease of use, the findings revealed data inconsistencies, gaps, and inaccuracies when compared to source documents. The study's scope was limited to PMTCT clinical evaluations in three health facilities, and budgetary constraints constrained the sample size. The researcher recommends further research comparing various data collection methodologies, especially in clinical evaluations, and suggests exploring the integration of AI or advanced technology to enhance character recognition for complex data types. While acknowledging the study's constraints, such as exclusive emphasis on PMTCT and a limited sample size, the research underscores the need for comprehensive investigations into the impact of different data collection methods on data quality in clinical settings, aiming to improve overall research practices and outcomes.