Abstract
Family health can be studied using the 1994-1995 National Health Interview Survey Disability Supplement by linking children to their mothers and other family members. However, the data item required to link is missing for 13% of children. We found that unlinked children and their probable mothers differed in many respects from their counterparts who could be linked, and exclusion of these mothers and their children from the analysis could bias results by introducing error due to incomplete coverage of the target population. We developed and validated a simple algorithm to match these children with their probable mother.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 73-86 |
| Number of pages | 14 |
| Journal | Research in Social Science and Disability |
| Volume | 3 |
| DOIs | |
| State | Published - 2003 |
| Externally published | Yes |
Fingerprint
ASJC Scopus subject areas
- Social Sciences (miscellaneous)
Cite this
THE IMPACT OF MISSING LINKAGE DATA IN FAMILY HEALTH RESEARCH : RESULTS FROM THE 1994-1995 NATIONAL HEALTH INTERVIEW SURVEY DISABILITY SUPPLEMENT. / Witt, Whitney P.; Riley, Anne W; Kasper, Judith D.
In: Research in Social Science and Disability, Vol. 3, 2003, p. 73-86.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - THE IMPACT OF MISSING LINKAGE DATA IN FAMILY HEALTH RESEARCH
T2 - RESULTS FROM THE 1994-1995 NATIONAL HEALTH INTERVIEW SURVEY DISABILITY SUPPLEMENT
AU - Witt, Whitney P.
AU - Riley, Anne W
AU - Kasper, Judith D.
PY - 2003
Y1 - 2003
N2 - Family health can be studied using the 1994-1995 National Health Interview Survey Disability Supplement by linking children to their mothers and other family members. However, the data item required to link is missing for 13% of children. We found that unlinked children and their probable mothers differed in many respects from their counterparts who could be linked, and exclusion of these mothers and their children from the analysis could bias results by introducing error due to incomplete coverage of the target population. We developed and validated a simple algorithm to match these children with their probable mother.
AB - Family health can be studied using the 1994-1995 National Health Interview Survey Disability Supplement by linking children to their mothers and other family members. However, the data item required to link is missing for 13% of children. We found that unlinked children and their probable mothers differed in many respects from their counterparts who could be linked, and exclusion of these mothers and their children from the analysis could bias results by introducing error due to incomplete coverage of the target population. We developed and validated a simple algorithm to match these children with their probable mother.
UR - http://www.scopus.com/inward/record.url?scp=34247442431&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34247442431&partnerID=8YFLogxK
U2 - 10.1016/S1479-3547(03)03005-7
DO - 10.1016/S1479-3547(03)03005-7
M3 - Article
AN - SCOPUS:34247442431
VL - 3
SP - 73
EP - 86
JO - Research in Social Science and Disability
JF - Research in Social Science and Disability
SN - 1479-3547
ER -