A METHOD FOR ESTIMATING YEAR OF BIRTH USING SOCIAL-SECURITY NUMBER
A METHOD FOR ESTIMATING YEAR OF BIRTH USING SOCIAL-SECURITY NUMBER
复制标题
DOI：
10.1093/oxfordjournals.aje.a113645
复制
发表时间：
1983-01-01
影响因子：
5
通讯作者：
SELTSER, RS
中科院分区：
文献类型：
--
作者：
BLOCK, G;MATANOSKI, GM;SELTSER, RS
描述了一种仅使用社会安全号码来估计出生年份的方法。该方法依赖于估计社会安全号码的发行年份，使用社会安全管理局表格（自1950年以来发行的号码）或外推法（1951年之前发行的号码）的精确信息。争议年龄是使用出生日期和社会安全号码已知的职业定义人群中的个人数据估计的。出生年份是出生年份减去年龄。观察到已知与估计出生年份的高度统计学显著相关性为0.91。预测在.+-内正确。在整个研究人群中，78%达到了5年。对于. apprx之后出生的人。1934年，预测。+-。正确年份的5年在95%的时间内实现，并且在±. 2年77%的时间。该方法可能是有用的流行病学研究的情况下，出生日期是失踪。一些应用进行了讨论。
A method for estimating year of birth using only Social Security number is described. The method relies on estimating the year of issue of the Social Security number, using either precise information from a Social Security Administration table (for numbers issued since 1950) or extrapolation (for numbers issued before 1951). Age at issue was estimated using data from individuals in an occupationally defined population for whom both birth date and Social Security number were known. Year of birth is then year of issue minus age at issue. A highly statistically significant correlation of 0.91 for known vs. estimated year of birth was observed. Predictions correct within .+-. 5 yr were achieved for 78% of the entire study population. For individuals born after .apprx. 1934, predictions within .+-. 5 yr of the correct year were achieved 95% of the time, and within .+-. 2 yr 77% of the time. The method may be useful in epidemiologic research in situations in which date of birth is missing. Some applications are discussed.