Psychometrics of Housing Quality Measurement in the American Housing Survey

AbstractThis article assesses the test-retest reliability, internal consistency, and convergent and predictive validity of the American Housing Survey inadequacy index. We find that the index does not appear to tap a single underlying construct of housing quality and does not differentiate among the worst quality units. We conclude that it may be urne to reconceptualize the elusive construct of housing quality.(ProQuest: ... denotes formulae omitted.)Data ShopData Shop, a department of Cityscape, presents short articles or notes on the uses of data in housing and urban research. Through this department, the Office of Policy Development and Research introduces readers to new and overlooked data sources and to improved techniques in using well-known data. The emphasis is on sources and methods that analysts can use in their own work. Researchers often run into knotty data problems involving data interpretation or manipulation that must be solved before a project can proceed, but they seldom get to focus in detail on the solutions to such problems. If you have an idea for an applied, data-centric note of no more than 3,000 words, please send a one-paragraph abstract to david.a.vandenbroucke@hud.gov for consideration.IntroductionAs the most comprehensive source of data on the U.S. housing stock, the American Housing Survey (AHS) is relied on by policymakers, practitioners, and researchers seeking answers to questions about the conditions, costs, and myriad other attributes of the nation's housing. For those developing their own surveys, the AHS is also a source of housing questions. Some of the most prominent social science studies of the past two decades, including the Moving to Opportunity for Fair Housing Demonstration (Shroder, 2001), Welfare, Children, & Families, A Three-City Study (Winston et al., 1999), and the Fragile Families and Child Well-Being Study (Reichman et al., 2001), include questions that are strikingly similar if not identical to the AHS items on housing characteristics and conditions.Of particular interest to many users is the AHS composite measure of housing inadequacy available on the public use database. This measure combines 15 individual questionnaire items on housing conditions into an index, setting numerical thresholds for the presence or absence of physical deficiencies in the dwelling to distinguish among "adequate," "moderately inadequate," and "severely inadequate" units. Both the AHS and data users refer to this composite as AHS's "housing quality" measure.1 Numerous published articles include the AHS measure in their analyses (for example, Carter, 2011; Friedman and Rosenbaum, 2004; Khadduri, 2007; Ross, Shlay, and Picon, 2012), the measure plays a prominent role in the U.S. Department of Housing and Urban Development's Worst Case Needs reports (for example, HUD, 201 lb), and it is also included in the frequently cited Joint Center for Housing Studies' State ojthe Nauon's Housing reports (for example, JCHS, 2010) and by the Millenial Housing Commission (2002).Despite widespread reliance on the AHS inadequacy index by a broad audience of users, little is known about its reliability, internal consistency, and validity. These attributes are typically referred to as psychometric features, because these tests were originally developed to assess indicators within the purview of psychologists, such as cognitive achievement, attitudes, and personality (Nunnaly, 1978).Our goal in this article is to shed light on each of these psychometric properties of the AHS inadequacy index. This information will enable users to assess, for example, (1) if the inadequacy index differentiates among dwellings of different housing quality, (2) if respondents can reliably answer the questions used to create the 15-item index, and (3) if we can be reasonably confident that the index is a valid representation of housing quality. The next section presents the composite index and its distribution. …