Perla Rocco J, Provost Lloyd P
University of Massachusetts Medical School, Department of Quantitative Health Sciences, 55 Lake Ave North, ACC Building 7th Floor, Worcester, MA 01655, USA.
Qual Manag Health Care. 2012 Jul-Sep;21(3):169-75. doi: 10.1097/QMH.0b013e31825e8806.
Sampling plays a major role in quality improvement work. Random sampling (assumed by most traditional statistical methods) is the exception in improvement situations. In most cases, some type of "judgment sample" is used to collect data from a system. Unfortunately, judgment sampling is not well understood. Judgment sampling relies upon those with process and subject matter knowledge to select useful samples for learning about process performance and the impact of changes over time. It many cases, where the goal is to learn about or improve a specific process or system, judgment samples are not merely the most convenient and economical approach, they are technically and conceptually the most appropriate approach. This is because improvement work is done in the real world in complex situations involving specific areas of concern and focus; in these situations, the assumptions of classical measurement theory neither can be met nor should an attempt be made to meet them. The purpose of this article is to describe judgment sampling and its importance in quality improvement work and studies with a focus on health care settings.
抽样在质量改进工作中起着重要作用。随机抽样(大多数传统统计方法所假定的)在改进情形中是个例外。在大多数情况下,会使用某种类型的“判断样本”从系统中收集数据。不幸的是,判断抽样并未得到很好的理解。判断抽样依靠具有过程和主题知识的人员来选择有用的样本,以便了解过程性能以及随着时间推移变化所产生的影响。在许多情况下,当目标是了解或改进特定过程或系统时,判断样本不仅是最便捷、最经济的方法,从技术和概念上讲也是最合适的方法。这是因为改进工作是在现实世界的复杂情形中进行的,涉及特定的关注领域和重点;在这些情形中,经典测量理论的假设既无法满足,也不应试图去满足。本文的目的是描述判断抽样及其在质量改进工作和研究中的重要性,重点关注医疗保健环境。