Multiple Imputation for Combined-Survey Estimation with Incomplete Regressors in One but Not Both Surveys

Details

• Personal Author:
  Baker EH ; Ghosh-Dastidar B ; Nazarov Z ; Rendall MS ; Weden MM
• Description:
  Within-survey multiple imputation (MI) methods are adapted to pooled-survey regression estimation where one survey has more regressors, but typically fewer observations, than the other. This adaptation is achieved through (1) larger numbers of imputations to compensate for the higher fraction of missing values, (2) model-fit statistics to check the assumption that the two surveys sample from a common universe, and (3) specifying the analysis model completely from variables present in the survey with the larger set of regressors, thereby excluding variables never jointly observed. In contrast to the typical within-survey MI context, cross-survey missingness is monotonic and easily satisfies the missing at random assumption needed for unbiased MI. Large efficiency gains and substantial reduction in omitted variable bias are demonstrated in an application to sociodemographic differences in the risk of child obesity estimated from two nationally representative cohort surveys. [Description provided by NIOSH]
• Subjects:
  Cohort Studies Occupational Health Safety
• Keywords:
  Author Keywords: Combining Data Child Obesity Children Cohort Studies Model Fit Statistics Multiple Imputation Obesity Panel Surveys Sociological Factors Statistical Analysis
• ISSN:
  0049-1241
• Document Type:
  Text
• Funding:
  Grant
• Genre:
  Journal Article
• Place as Subject:
  Alabama ; California ; Indiana ; Maryland ; OSHA Region 3 ; OSHA Region 4 ; OSHA Region 5 ; OSHA Region 9
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Pages in Document:
  483-530
• Volume:
  42
• Issue:
  4
• NIOSHTIC Number:
  nn:20054110
• Citation:
  Sociol Methods Res 2013 Nov; 42(4):483-530
• Contact Point Address:
  Michael S. Rendall, University of Maryland, 3139 Art-Sociology Building, College Park, MD 20742, USA
• Email:
  mrendall@umd.edu
• Federal Fiscal Year:
  2014
• Performing Organization:
  RAND Corporation, Santa Monica, California
• Peer Reviewed:
  True
• Start Date:
  20080601
• Source Full Name:
  Sociological Methods & Research
• End Date:
  20110531
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:4ce54f5db290f3de33d71898a2f864d6de68570e133e509b82ee20dfadd455385681cccbcaf68854c3b4f88356b6248aaf731708b7932b147f63ceb628e1c0a8
• Download URL:
  https://stacks.cdc.gov/view/cdc/216744/cdc_216744_DS1.pdf
• File Type:
  [PDF - 313.89 KB ]