112
ent kinds of housing deficiencies. Unfortunately,
AHS data is not useful for local communities or
neighborhoods because it is limited to specific
metropolitan areas.
From the beginning of its implementation,
scholars and analysts began to use the AHS information in different ways and for different purposes. Early analysts began to measure inadequate
housing by creating their own AHS-based definitions (see Levine 1978). Simonson (1981) records
a detailed narrative of early studies and their definitions using indicators from the AHS. What is
relevant in these early efforts using AHS data is
that depending on the indicators used to conceptualize inadequate housing, the number of units
could increase or decrease, again proving the
effect of the changing nature of the concept.
In addition to the efforts to define and measure
inadequate housing using AHS indicators, other
scholars began to use the data to assess the statistical relevance of individual indicators. For
instance, Newman and Struyk (1983), in their
study addressing the relationship between poverty and housing deprivation, relied on AHS’s
definition. Newman and Schnare (1988), also
using AHS information and definitions, examined the relationship between income and housing assistance programs. As an example of the
combination of sources of indicators, Bianchi
et al. (1982) examined the degree to which racial
differences in housing narrowed over time using
1960 Census information and AHS data.
Scholars have also used AHS data to test the
validity of alternative measures of inadequate
housing (e.g., housing inspection assessments)
by comparing them against the classification
used by HUD. For instance, using information on
the physical conditions of housing units in the
Experimental Housing Allowance Program
(EHAP), Budding (1980) found a higher number
of inadequate housing units compared to estimates using AHS data.
Similar to Weicher’s (1979) findings about the
problem of definition and the apparent decrease
in inadequate housing, using US Census housing
information, two more recent studies found
inconsistencies in the accuracy of the indicators
used to identify inadequate housing as conceptualized by the AHS. In a work prepared for HUD,
Eggers and Moumen (2013) found a decline in
the presence of inadequate housing after the definition was instituted by the AHS. The authors
also express a concern for how this apparent
improvement has decreased interest in the topic.
This lack of urgency has been reinforced by the
apparent decline in the proportion of severely
inadequate housing among low-income households (Orr and Peach 1999).
Empirical work by Emrath and Taylor (2012)
found not only that the AHS’s indicators of physically inadequate housing units are not significant
but they also had the wrong signs. The authors
propose instead new measurements of housing
inadequacy by using other indicators contained
in the AHS. The proposed new definition reveals
a greater number of housing units in the United
States as being physically inadequate. Emrath
and Taylor’s new approach of using only indicators of external physical conditions is the
approach taken in this study for the creation of
the housing quality composite indicator.
Scholarly work using AHS data is abundant.
But although the AHS information is exhaustive
and contains data for a good period time regarding the same housing units, this is limited to specific metropolitan areas, making it not useful for
analysis of other geographical areas. With this
limitation from the AHS in mind, other scholars
began to rely on other data sources, specifically
on the implementation of housing assessment
surveys. A notable study of this type is the work
by Kain and Quigley (1970), who relied on 39
variables provided by three separate surveys for
the City of St. Louis. The concerns with these
types of surveys or assessments lie both in terms
of time and cost of implementation and the potential errors made by collectors in the field.
Considering the geographical limitations of
official information and the disadvantages of
lengthy surveys, scholars have attempted to come
up with different strategies using alternative indicators as a more efficient way to identify inadequate housing. Relying on the latest empirical
work on determinants of quality of housing,
L. Estevez
ent kinds of housing deficiencies. Unfortunately,
AHS data is not useful for local communities or
neighborhoods because it is limited to specific
metropolitan areas.
From the beginning of its implementation,
scholars and analysts began to use the AHS information in different ways and for different purposes. Early analysts began to measure inadequate
housing by creating their own AHS-based definitions (see Levine 1978). Simonson (1981) records
a detailed narrative of early studies and their definitions using indicators from the AHS. What is
relevant in these early efforts using AHS data is
that depending on the indicators used to conceptualize inadequate housing, the number of units
could increase or decrease, again proving the
effect of the changing nature of the concept.
In addition to the efforts to define and measure
inadequate housing using AHS indicators, other
scholars began to use the data to assess the statistical relevance of individual indicators. For
instance, Newman and Struyk (1983), in their
study addressing the relationship between poverty and housing deprivation, relied on AHS’s
definition. Newman and Schnare (1988), also
using AHS information and definitions, examined the relationship between income and housing assistance programs. As an example of the
combination of sources of indicators, Bianchi
et al. (1982) examined the degree to which racial
differences in housing narrowed over time using
1960 Census information and AHS data.
Scholars have also used AHS data to test the
validity of alternative measures of inadequate
housing (e.g., housing inspection assessments)
by comparing them against the classification
used by HUD. For instance, using information on
the physical conditions of housing units in the
Experimental Housing Allowance Program
(EHAP), Budding (1980) found a higher number
of inadequate housing units compared to estimates using AHS data.
Similar to Weicher’s (1979) findings about the
problem of definition and the apparent decrease
in inadequate housing, using US Census housing
information, two more recent studies found
inconsistencies in the accuracy of the indicators
used to identify inadequate housing as conceptualized by the AHS. In a work prepared for HUD,
Eggers and Moumen (2013) found a decline in
the presence of inadequate housing after the definition was instituted by the AHS. The authors
also express a concern for how this apparent
improvement has decreased interest in the topic.
This lack of urgency has been reinforced by the
apparent decline in the proportion of severely
inadequate housing among low-income households (Orr and Peach 1999).
Empirical work by Emrath and Taylor (2012)
found not only that the AHS’s indicators of physically inadequate housing units are not significant
but they also had the wrong signs. The authors
propose instead new measurements of housing
inadequacy by using other indicators contained
in the AHS. The proposed new definition reveals
a greater number of housing units in the United
States as being physically inadequate. Emrath
and Taylor’s new approach of using only indicators of external physical conditions is the
approach taken in this study for the creation of
the housing quality composite indicator.
Scholarly work using AHS data is abundant.
But although the AHS information is exhaustive
and contains data for a good period time regarding the same housing units, this is limited to specific metropolitan areas, making it not useful for
analysis of other geographical areas. With this
limitation from the AHS in mind, other scholars
began to rely on other data sources, specifically
on the implementation of housing assessment
surveys. A notable study of this type is the work
by Kain and Quigley (1970), who relied on 39
variables provided by three separate surveys for
the City of St. Louis. The concerns with these
types of surveys or assessments lie both in terms
of time and cost of implementation and the potential errors made by collectors in the field.
Considering the geographical limitations of
official information and the disadvantages of
lengthy surveys, scholars have attempted to come
up with different strategies using alternative indicators as a more efficient way to identify inadequate housing. Relying on the latest empirical
work on determinants of quality of housing,
L. Estevez