Nielsen-Thompson N, Boyer E M
Dental Hygiene Program, University of Iowa, Iowa City, USA.
J Dent Hyg. 1994 Jul-Aug;68(4):188-92.
This study was designed to investigate the reporting of nonresponse in mail survey research and to determine if nonresponse as a result of inaccurate addresses varied by sampling frame. Nonresponse bias is defined as researchers' inability to contact some members of a sample, or the refusal of some sample members to provide information.
Published survey research articles in the Journal of Dental Hygiene from 1988 through 1992 were reviewed for nonresponse information. Thirty-nine articles described studies that met the requirements of being a mail survey of dental hygiene subjects. All 39 articles included information about the population base and the number of subjects in the sample. Descriptive statistics were used to analyze the data.
Response rates ranging from 33% to 92% were reported in 37 articles. Only 15 of the 37 articles included information on the ability to contact subjects. The percentage of subjects who could not be contacted ranged from 1.1% to 13.3%. Professional organization membership rosters provided the most accurate mailing lists, followed by school class rosters. Least accurate were licensure lists and magazine lists.
More attention to research design may help dental hygiene researchers recognize the impact of nonresponse bias.
本研究旨在调查邮寄调查研究中无应答情况的报告，并确定因地址不准确导致的无应答是否因抽样框架而异。无应答偏差被定义为研究人员无法联系到样本中的某些成员，或某些样本成员拒绝提供信息。
对1988年至1992年发表在《牙科保健杂志》上的调查研究文章进行回顾，以获取无应答信息。39篇文章描述了符合牙科保健受试者邮寄调查要求的研究。所有39篇文章都包含了关于总体基数和样本中受试者数量的信息。使用描述性统计分析数据。
37篇文章报告的应答率在33%至92%之间。37篇文章中只有15篇包含了联系受试者能力的信息。无法联系到的受试者百分比在1.1%至13.3%之间。专业组织成员名单提供的邮寄名单最准确，其次是学校班级名单。最不准确的是执照名单和杂志名单。
更多地关注研究设计可能有助于牙科保健研究人员认识到无应答偏差的影响。
