Missing Data in Quantitative Social Research

Almost invariably, the data available to the social scientist display one or more characteristics of missing information. Even though reasons for non response are varied, most frequently, they reflect the unwillingness of respondents to provide information on undesirable social behaviours and on issues considered as private. Besides these, sloppy research designs often leads to ambiguous and poorly structured survey questions which provide a recipe for low response. Longitudinal surveys also suffer from incompleteness due to attrition resulting from death and emigration, while in retrospective surveys, memory effect might be a major source of non-response. While there is no consensus among methodologists on the single most effective technique of handling missing information, certain pertinent questions need to be addressed: should we completely ignore the missing data and proceed with the analysis? What are the implicit assumptions if one adopts such an approach and how unbiased will our estimates be? This paper reviews a variety of methods of handling missing information.

[1]  D. Rubin,et al.  Multiple Imputation for Interval Estimation from Simple Random Samples with Ignorable Nonresponse , 1986 .

[2]  Donald B. Rubin,et al.  Multiple Imputation of Industry and Occupation Codes in Census Public-use Samples Using Bayesian Logistic Regression , 1991 .

[3]  Ingram Olkin,et al.  Incomplete data in sample surveys. Vol. 1: report and case studies , 1983 .

[4]  R. Fay.  Causal Models for Patterns of Nonresponse , 1986 .

[5]  D. Rubin.  Formalizing Subjective Notions about the Effect of Nonrespondents in Sample Surveys , 1977 .

[6]  S. F. Buck.  A Method of Estimation of Missing Values in Multivariate Data Suitable for Use with an Electronic Computer , 1960 .

[7]  Y. Haitovsky.  Missing Data in Regression Analysis , 1968 .

[8]  P. Horst.  The missing data matrix. , 1968, Journal of clinical psychology.

[9]  Y. Dodge.  Analysis of Experiments with Missing Data , 1985 .

[10]  D. Rubin,et al.  Statistical Analysis with Missing Data , 1988 .

[11]  Allen S. Mandel.  Comment … , 1978, British heart journal.

[12]  Roger L. Brown.  Efficacy of the indirect approach for estimating structural equation models with missing data: A comparison of five methods , 1994 .

[13]  R. Fay.  Alternative Paradigms for the Analysis of Imputed Survey Data , 1996 .

[14]  Zvi Griliches,et al.  ECONOMIC DATA ISSUES , 1986 .

[15]  Joseph Sedransk,et al.  A Bayesian Procedure for Imputing Missing Values in Sample Surveys , 1986 .

[16]  William S. Reece,et al.  Imputation of Missing Values When the Probability of Response Depends on the Variable Being Imputed , 1982 .

[17]  Roderick J. A. Little,et al.  Statistical Analysis with Missing Data , 1988 .

[18]  D. Rubin.  Multiple Imputation After 18+ Years , 1996 .

[19]  Roger A. Sugden,et al.  Multiple Imputation for Nonresponse in Surveys , 1988 .

[20]  M. Glasser,et al.  Linear Regression Analysis with Missing Observations among the Independent Variables , 1964 .

[21]  Jae-On Kim,et al.  The Treatment of Missing Data in Multivariate Analysis , 1977 .

[22]  Ingram Olkin,et al.  Incomplete data in sample surveys , 1985 .

[23]  M. Woodbury.  A missing information principle: theory and applications , 1972 .

[24]  J. Rao.  On Variance Estimation with Imputed Survey Data , 1996 .

[25]  Roderick J. A. Little,et al.  The Analysis of Social Science Data with Missing Values , 1989 .

[26]  G. Kalton,et al.  The treatment of missing survey data , 1986 .

[27]  J. Shao,et al.  Jackknife variance estimation with survey data under hot deck imputation , 1992 .

[28]  Werner Vach,et al.  Logistic Regression with Missing Values in the Covariates , 1994 .