Improving the quality of data collection in midwifery research: the role of the reliability study

No Thumbnail Available
Authors
Knight, Bridget
Shields, Beverley
Turner, Martina
Hattersley, Andrew
Powell, Roy
Journal
Evidence-Based Midwifery
Type
Journal Article
Publisher
Royal College of Midwives
DOI
Rights
COPYRIGHT 2005 Redactive Media Group
Abstract Aim. To establish the importance of data quality in measurements collected by midwives in both the research and the routine context. Objectives. To identify the role of reliability studies in ensuring data quality in a large-scale epidemiological study; and to discuss the implications for routine data collection in clinical practice. Method. Detailed anthropometric measurements were taken by several midwives from families involved with the Exeter Family Study of Childhood Health (EFSOCH). Protocols and procedures were developed to reduce the error component of measurements taken, and to assess the levels of agreement between the midwives. Findings/results. The authors identified both random and systematic errors made by the midwives. With the development of clear, written protocols, and by spending time and effort on practice, it was possible to obtain acceptable levels of agreement between midwives to enable collective data analysis. Implications. In evaluating published work, midwives must be aware of the need to identify what steps were undertaken to ensure quality of the data that analysis is based on. In clinical practice, midwives should be aware of the need for accuracy in the collection of even the most routine data. When taking anthropometric measures on newborns or neonates, it is necessary to use standard methodology and spend time to ensure accuracy of measurement.
Citation
Note
The article is available via Open Access. Click on the 'Additional link' above to access the full-text.