Smith Anthony M A, de Visser Richard O, Rissel Chris E
Australian Research Centre in Sex, Health and Society, La Trobe University, Melbourne, Australia.
Ann Epidemiol. 2005 Mar;15(3):232-5. doi: 10.1016/j.annepidem.2004.06.003.
To compare the estimates of HIV-related risk derived from a sample drawn through random digit dialing and that component of the sample drawn from households listed in a telephone directory.
The prevalence of key outcomes, and their 95% confidence intervals, was estimated for the total sample and for that component of the sample drawn from households listed in a telephone directory.
On all outcome measures the sample derived from listed households was more conservative. With few exceptions, the estimates derived from the overall sample and from listed households were not significantly different.
A continued reliance on random digit dialing seems unwarranted.
比较通过随机数字拨号抽取的样本以及从电话簿中列出的家庭抽取的样本部分所获得的与艾滋病病毒相关风险的估计值。
对总样本以及从电话簿中列出的家庭抽取的样本部分，估计关键结果的患病率及其95%置信区间。
在所有结果指标上，来自列出家庭的样本更为保守。除少数例外，总体样本和列出家庭的估计值没有显著差异。
继续依赖随机数字拨号似乎没有必要。
