How many is enough? Determining optimal sample sizes for normative studies in pediatric neuropsychology
How many is enough? Determining optimal sample sizes for normative studies in pediatric neuropsychology
复制标题
DOI：
10.1080/09297040701233875
复制
发表时间：
2007-11-01
影响因子：
2.2
通讯作者：
Holler, Karen A.
中科院分区：
文献类型：
--
作者：
Bridges, Ana J.;Holler, Karen A.
本研究的目的是确定儿童神经心理学规范的置信区间（CI）如何随样本量的变化而变化，并确定规范性研究的最佳样本量。首先，作者计算了四种常用神经心理学工具的一组已发表的儿科规范的95% CI。其次，计算不同样本量（n = 5至n = 500）的95% CI。结果表明，一些儿科规范具有不可接受的宽CI，规范性研究最好每个细胞使用50至75名参与者。较小的样本量可能导致过度病理化的结果，而获得较大样本的成本可能不合理。
The purpose of this investigation was to determine how confidence intervals (CIs) for pediatric neuropsychological norms vary as a function of sample size, and to determine optimal sample sizes for normative studies. First, the authors calculated 95% CIs for a set of published pediatric norms for four commonly used neuropsychological instruments. Second, 95% CIs were calculated for varying sample size (from n = 5 to n = 500). Results suggest that some pediatric norms have unacceptably wide CIs, and normative studies ought optimally to use 50 to 75 participants per cell. Smaller sample sizes may lead to over-pathologizing results, while the cost of obtaining larger samples may not be justifiable.