Session: Introducing Missing Data Analysis in Social Work Research (Society for Social Work and Research 21st Annual Conference - Ensure Healthy Development for all Youth)

178 Introducing Missing Data Analysis in Social Work Research

Schedule:
Saturday, January 14, 2017: 8:00 AM-9:30 AM
Mardi Gras Ballroom B (New Orleans Marriott)
Cluster: Research Design and Measurement
Speakers/Presenters:
Ding Gen("Din") Chen, PhD, University of North Carolina at Chapel Hill, Shiyou Wu, MSW, University of North Carolina at Chapel Hill and Qi Wu, PhD, University of Mississippi
Background: Missing data are near universal in quantitative social work research. Almost all social work researchers have faced the problems of missing data at some point. However, not all the social work researchers assessed missingness or used appropriate ways to deal with the missing data. Instead, researchers often drop the missing values (e.g., listwise deletion), which reduces the sample size, lowers statistical power, and introduces the possibility of biased parameter estimations. Such inefficient statistical inference would lead to erroneous research conclusions and policy recommendations. Purpose: This workshop aims to address the problems of missing data in social work research, and to share three different missing data mechanisms or typologies including missing completely at random (MCAR), missing at random (MAR) and not missing at random (NMAR). Moreover, this workshop will introduce how to conduct two commonly-used model-based methods for missing data analysis including the multiple imputation (Little & Rubin, 2002; Reiter & Raghunathan, 2007) and maximum likelihood (Allison, 2012).

Contents: This workshop will: (a) review the different missing data mechanisms (i.e., MCAR, MAR and NMAR); (b) introduce the multiple imputation and maximum likelihood methods in regression modeling and discuss the advantages and disadvantages of each method; and (c) illustrate using R (free software) and Stata (commonly used in social work research) to analyze real data from social work studies and compare the consistence from different software.

Pedagogical Techniques: This workshop will use a PowerPoint presentation to introduce the statistical principles of missing data in regression analysis, and will show to the participants how to conduct the multiple imputation and maximum likelihood in R/Stata with real data examples from social work research. Handouts with R/Stata syntax of running multiple imputation and maximum likelihood will be provided to the participants.  

Significance:  This workshop aims to address the missing data problem in social work research by reviewing the existing statistical methods. Real data analysis using R/Stata will be illustrated. The proposed approach will help to advance scientific knowledge, improve the rigorousness of quantitative research in social work areas, and provide more accurate guidance for social work practice.

See more of: Workshops