Missing Data Analysis

Dr. Paul Allison

Aim of Course:

Conventional methods for handling missing data, like listwise deletion or regression imputation, waste data, sacrifice power, and can yield biased estimates of parameters and standard errors. Much better results can be obtained with the newer methods of maximum likelihood or multiple imputation, which have become practical in the last few years with the introduction of widely available and user friendly software. They have very similar statistical properties. If the assumptions are met, they are approximately unbiased and efficient--that is, they have minimum sampling variance. What's remarkable is that these newer methods depend on less demanding assumptions than those required for conventional methods for handling missing data. At present, maximum likelihood is best suited for linear models or log-linear models for contingency tables. Multiple imputation, on the other hand, can be used for virtually any statistical problem.

This course will cover the theory and practice of both maximum likelihood and multiple imputation.

Who Should Take This Course:

Any statistical analyst who works with multivariate data is likely to encounter missing observations and will benefit from this course.

For those enrolled in Professional Advancement Programs, this is a required or elective course in the following Programs:

• Biostatistics (epidemiology) - elective
• Biostatistics (controlled trials) - elective
• Data Mining - elective
• Statistics for Social Sciences - elective
• Statistics for Environmental Science - elective

Course Program:

The course is structured as follows

The Instructor:

Paul Allison, Professor of Sociology at the University of Pennsylvania. He is the author of Missing Data (Sage 2001), Multiple Regression: A Primer (Pine Forge 1999), Survival Analysis Using SAS: A Practical Guide (SAS Institute 1995), Event History Analysis (Sage 1984), several other books, and numerous articles on regression analysis, log-linear analysis, logit analysis, latent variable models, missing data, and inequality measures. A former Guggenheim Fellow, he is also on the editorial board of Sociological Methods and Research. In 2001 he received the Paul Lazarsfeld Memorial Award for Distinguished Contributions to Sociological Methodology.

Organization of the Course:

The course takes place over the internet, at statistics.com. During each course week, you participate at times of your own choosing - there are no set times when you must be online. Course participants will be given access to a private discussion board. In class discussions led by the instructor, you can post questions, seek clarification, and interact with your fellow students and the instructor. The course is scheduled to take place over 4 weeks, and typically requires 10-15 hours per week. At the beginning of each week, you receive the relevant material, in addition to answers to exercises from the previous session. During the week, you are expected to go over the course materials and work through exercises. Discussion among participants is encouraged. The instructor will provide answers and comments.

Certificates and Grades:

You may be interested only in learning the material presented, and not be concerned with grades or certificates. Or you may be enrolled in a statistics.com Professional Advancement Program that requires demonstration of proficiency in the subject, in which case your work will be assessed for purposes of issuing a grade. Or you may require only a "Certificate of Course Completion," along with professional development credit in the form of Continuing Education Units (CEU's). As you begin the class, you will be asked to specify your category.

Credit:

This course offers continuing education units (CEU's). For those successfully completing the course (generally this means marks of 50% or better on the homework), 5.0 CEU's and a certificate will be issued by statistics.com, upon request.

Dates:

Nov. 21 - Dec. 19, 2008
Click here to be notified of future course offerings.

Participants gain access to the online materials on the first day of the course, and typically spend about 10-15 hours per week (at their convenience). You retain full access to course materials, including discussion board, for two weeks after the course closing date.

Level:

intermediate/advanced

Prerequisite:

To take this course, you should have a good working knowledge of the principles and practice of multiple regression, as well as elementary statistical inference. But you do not need to know matrix algebra, calculus, or likelihood theory.

Course Text:

The required text is Missing Data, by Paul Allison (Sage, 2001). It can be ordered directly from the publisher here. Sage Publication offers discounts to students at statistics.com for many of their titles when the code S06SC is used during checkout on their website (the 0 is a zero not an alphabetical O).

Software:

Maximum likelihood for linear models will be demonstrated with Amos 6 (students may use AMOS 5 demo version for exercises), a package designed for estimating structural equation models with latent variables, and with LEM, a freeware package for categorical data analysis. Multiple imputation will be demonstrated with the MI procedure in SAS and the ICE command in Stata. For information on obtaining AMOS, LEM, and other software, click here.

A note about software use in the exercises: All of Session 3's exercises, and several other exercises, require SAS. Course participants without SAS on their computer should seek access to a SAS installation, particularly for the third week. There is still lots to do in the course without using SAS

Registration:

Register Online - $449
Register Online (academic) - $349 (you must be affiliated with a college, university or high school)

Add $50 service fee if you require a prior invoice, or if you need to submit a purchase order or voucher, pay by wire transfer or EFT, or refund and reprocess a prior payment. Please use this printed registration form, for these and other special orders.

Note: Courses may fill up at any time and registrations are processed in the order in which they are received. Your registration will be confirmed for the first available course date, unless you specify otherwise.