Ronaldo Iachan, PhD
ICF International, Calverton, Maryland
Dr. Iachan, a senior statistician, has 30 years of experience in statistical methods and applications, particularly in the areas of survey sampling design and analysis. For ICF, Dr. Iachan provides statistical expertise across divisions in projects in the areas of health, education, and social studies. He was a professor at Iowa State University and at the University of Wisconsin–Madison. He has served on ICF’s IRB for 12 years, and has been a statistical editor for the Journal of the American Medical Association for the past 15 years. He has extensive experience in statistical design and analysis, with more than 30 refereed articles published in statistical methods,. Dr. Iachan has extensive experience providing sampling and survey design support for many cancer-related projects for the Centers for Disease Control and Prevention (CDC) and other agencies, including cancer registry studies and heart disease prevention.
502 – Innovative Statistical Methods for Complex Survey Data
Are Utility Subsidies Accurately Predicting Assisted Households’ Actual Utility Expenditures?: A Methodological Examination
Davia Spado
ICF International
Lee Harding
ICF International
Ronaldo Iachan, PhD
ICF International, Calverton, Maryland
Households receiving rental subsidies through the U.S. Department of Housing and Urban Development's (HUD) assisted housing programs receive a reduction in their monthly rent, or a utility allowance (UA), for out-of-pocket utility costs. UA amounts are estimated by housing sites and are not necessarily equal to a household's actual monthly utility expenses. They are derived using a variety of methods, including the HUD Utility Schedule Model (HUSM). The Utility Allowance Comparison study, sponsored by HUD, seeks to ascertain whether the HUSM UA method is better or worse at predicting actual out-of-pocket utility expenses than a comparison UA calculated via other methods. We use data collected from a nationally representative sample and mean square errors (MSEs) to assess the accuracy of the two UA methods in predicting the actual utility expenditure. The analysis uses jackknife replication to produce standard errors and confidence intervals of the two MSEs, and a determination of prediction performance between the methods is made. Results may inform policies and methods by which UAs are calculated to mitigate both under- and over-subsidizing for out-of-pocket utility costs.