Detecting and Statistically Correcting Sample Selection Bias

ABSTRACT Researchers seldom realize 100% participation for any research study. If participants and non-participants are systematically different, substantive results may be biased in unknown ways, and external or internal validity may be compromised. Typically social work researchers use bivariate tests to detect selection bias (e.g., χ2 to compare the race of participants and non-participants). Occasionally multiple regression methods are used (e.g., logistic regression with participation/non-participation as the dependent variable). Neither of these methods can be used to correct substantive results for selection bias. Sample selection models are a well-developed class of econometric models that can be used to detect and correct for selection bias, but these are rarely used in social work research. Sample selection models can help further social work research by providing researchers with methods of detecting and correcting sample selection bias.

[1]  Christopher Winship,et al.  THE ESTIMATION OF CAUSAL EFFECTS FROM OBSERVATIONAL DATA , 1999 .

[2]  M. Smyer,et al.  Response Bias Using Two-Stage Data Collection , 1988 .

[3]  W. Shadish,et al.  Experimental and Quasi-Experimental Designs for Generalized Causal Inference , 2001 .

[4]  D. Relles,et al.  Tools for intuition about sample selection bias and its correction , 1997 .

[5]  Julia H. Littell,et al.  Effects of the duration, intensity, and breadth of family preservation services: A new analysis of data from the illinois family first experiment☆ , 1997 .

[6]  J. Heckman.  The Common Structure of Statistical Models of Truncation, Sample Selection and Limited Dependent Variables and a Simple Estimator for Such Models , 1976 .

[7]  Christopher Winship,et al.  Models for Sample Selection Bias , 1992 .

[8]  James J. Heckman,et al.  Assessing the Case for Social Experiments , 1995 .

[9]  John W. Polak.  Empirical Analysis of Attrition and Underreporting in Mailback and Personal Interview Panel Surveys , 1999 .

[10]  R. Berk.  An introduction to sample selection bias in sociological data. , 1983 .

[11]  David W. Wright,et al.  Detecting and Correcting Attrition Bias in Longitudinal Family Research , 1995 .

[12]  V. Bengtson,et al.  Geographic distance and contact between middle-aged children and their parents: the effects of social class over 20 years. , 1997, The journals of gerontology. Series B, Psychological sciences and social sciences.

[13]  S. Myers,et al.  The effects of sample selection bias on racial differences in child abuse reporting. , 1998, Child abuse & neglect.

[14]  Thomas P. Vartanian,et al.  Adolescent Neighborhood Effects on Labor Market and Economic Outcomes , 1999, Social Service Review.

[15]  T. Cook,et al.  Quasi-experimentation: Design & analysis issues for field settings , 1979 .

[16]  I. Garfinkel,et al.  Absent Father's Ability to Pay More Child Support , 1990 .

[17]  E. Foster,et al.  An Evaluator's Guide To Detecting Attrition Problems , 1996 .

[18]  J. Heckman.  Dummy Endogenous Variables in a Simultaneous Equation System , 1977 .

[19]  P J Gruenewald,et al.  Sample selection bias in the emergency room: an examination of the role of alcohol in injury. , 1998, Addiction.

[20]  A. Kaylor.  The Effect of Initial Placement into Kinship Foster Care on Reunification from Foster Care , 2001 .

[21]  Lawrence C. Marsh,et al.  Sample Selection Bias Correction for Missing Response Observations , 2000 .

[22]  A. Reynolds,et al.  Quasi-Experimental Estimates of the Effects of a Preschool Intervention , 1995 .

[23]  Petra E. Todd,et al.  Matching As An Econometric Evaluation Estimator: Evidence from Evaluating a Job Training Programme , 1997 .

[24]  W. Greene.  Sample Selection Bias as a Specification Error: Comment , 1981 .

[25]  Robert A. Moffitt,et al.  Program Evaluation With Nonexperimental Data , 1991 .

[26]  I. Piliavin,et al.  Transitions from and Returns to Out-of-Home Care , 1997, Social Service Review.

[27]  John G. Orme,et al.  Introduction to multiple regression for categorical and limited dependent variables , 2001 .

[28]  James D. Wright,et al.  Evaluating an alcohol and drug treatment program for the homeless: An econometric approach , 1997 .

[29]  D. Bybee,et al.  Tracking and Follow-Up Methods for Research On Homelessness , 1993 .

[30]  H. Wiegand,et al.  Kish, L.: Survey Sampling. John Wiley & Sons, Inc., New York, London 1965, IX + 643 S., 31 Abb., 56 Tab., Preis 83 s. , 1968 .

[31]  Kosuke Imai,et al.  Survey Sampling , 1998, Nov/Dec 2017.

[32]  L. Delbeke.  Quasi-experimentation - design and analysis issues for field settings - cook,td, campbell,dt , 1980 .

[33]  R. Barth,et al.  Adult transracial and inracial adoptees: effects of race, gender, adoptive family structure, and placement history on adjustment outcomes. , 1999, The American journal of orthopsychiatry.

[34]  John G. Orme,et al.  Parental and familial characteristics of family foster care applicants , 2004 .

[35]  S. Semaan,et al.  Effects of Intervention Attrition and Research Attrition on the Evaluation of an HIV Prevention Program , 1996 .

[36]  Jiming Jiang,et al.  GENDER BIAS AND THE COLLEGE PREDICTIONS OF THE SATS: A Cry of Despair , 1999 .

[37]  Ron D. Hays,et al.  Adjusting for Attrition in School-Based Samples , 1997 .

[38]  J. Heckman.  Sample selection bias as a specification error , 1979 .

[39]  David A. Hensher,et al.  ISSUES IN THE PRE-ANALYSIS OF PANEL DATA , 1987 .