Calhoun-3267-19.

qxd

3/29/2005

2:30 PM

Page 327

19
Poverty and Life Chances:
The Conceptualization and
Study of the Poor
D A LT O N C O N L E Y

INTRODUCTION: POVERTY AMID PLENTY

This chapter is concerned with the theoretical
conceptualization of poverty in rich, developed
countries and the estimation of its effects on
offspring. The difficulties of conceptualizing
poverty amid plenty are perhaps best illustrated
by a speech given by a member of the Forbes
400 richest Americans, Thomas Monaghan,
who himself rose to great wealth from meager
origins. To me one of the most exciting things
in the world is being poor, he began his lecture.
To explain what he meant, Monaghan cited a
study that concluded that a family of four could
survive on $68 per year back around 1970
(which would make it $256 today). Now youre
probably wondering how you can live on $68 a
year. The first thing you do is go to the Farm
Bureau and buy a hundred-pound bag of powdered milk While youre at the Farm Bureau
you buy yourself a bushel of oats or wheat or
corn, and you mash that stuff up And you
grow some vegetables and you get a few vitamin pills to supplement your diet. And I think
thats exciting. He went on to talk about how
cheaply he lived in a house trailer, calling it the

greatest living I ever did. He concluded his

speech with a rhetorical appeal: Oh gosh,
Monaghan said, Id love to talk to all these
people who say they cant get by.1
We could debate the actual numbers that
is, exactly how cheaply someone could survive
in the contemporary United States or a similarly developed country and we could question the hypocrisy of a man worth hundreds of
millions of dollars castigating the poor for
their implied whining, but that would miss the
deeper point that Monaghan raises. Namely,
what does it mean to be poor in a country
when starvation and death from the elements
is rare? This question inevitably leads us to the
debate over absolute versus relative measures
of poverty. As we will see, Monaghans reasoning is not that far afield from a long tradition
of absolute poverty measurement that has
based its calculations on the cost of food.
Since the end of the eighteenth century,
many individuals and institutions have tried to
come up with the magical perfect absolute measure of poverty. This led theorists to attempt to
quantify the basic necessities needed to live;
more to the point, they tried to define poverty

Calhoun-3267-19.qxd

328

3/29/2005

2:30 PM

Page 328

THE SAGE HANDBOOK OF SOCIOLOGY

in terms of food requirements. Specifically, a

household fell into poverty if its income fell
below the necessary level to purchase food to
physically sustain itself.2 Attempts to establish
such a minimal standard started in England in
1795 when the town of Speenhamland instituted a relief program that made up the difference between a workers wage and the cost of
bread sufficient to feed him and his family
(Stone, 1994: 85). In 1901, Rowntree attempted
to devise a specific measure in York, England
when he documented an income level below
which the necessities to maintain ones physical
efficiency could not be afforded. His approach
was institutionalized in the United Kingdom by
the Beveridge Report in 1942.
The most famous American rendition of the
food-based measurement of poverty status was
that of Mollie Orshansky in her 1963 article
Children of the Poor. To estimate the poverty
line, she used a strategy not unlike that implicit
in Monaghans speech. She took the US
Department of Agricultures recommendations for the minimum amount of healthy
food, estimated the cost for a variety of family
types (62 in all) and multiplied this figure by a
factor of three.3 Soon, this became the official
poverty line of the United States. As such, it has
been the definition of poverty that has most
frequently served as the straw man against which
researchers have suggested alternatives.
Orshansky has been assailed from all sides for
the choices she made in drawing the line in the
United States. Early criticism revolved around
her choice of three as the multiplier. Rose
Friedman, for example, argued in 1965 that
three was too high since the poor often spent a
greater proportion of their income on food
during this time period (the actual figure being
closer to 60 per cent, according to Friedman).
Friedmans estimation lowered the poverty
threshold substantially and halved the number
of individuals living in poverty at the time.
However, Friedmans argument appears flawed
because of its circularity: namely, the poor may
have been spending more of their resources on
food since they were poor. In other words, since
food is the most basic necessity of all, we do
not know what other necessities the poor may
have been forsaking due to the fact that they

were spending over half their money to keep

themselves fed (such as medical care, adequate
shelter, and so on). Alternatively, Alan Haber
(1966) argued that Orshanskys survey data
from the 1950s overestimated the percentage of
family income spent on food in the 1960s, suggesting that it had fallen to one-fourth, as illustrated by the 19601 Consumer Expenditure
Survey. The percentage of family income spent
on food has steadily dropped since this period
as well. Now housing makes up a much larger
proportion of household budgets.4
Another conservative line of argument for
change is to adjust what we call total family
income. Some analysts have argued that total
family income is a weak measure of the consumptive power of the poor since many receive
in-kind benefits such as food stamps and
Medicaid that raise their standard of living
but which are not counted as income. Liberals
counter that if we include in-kind benefits for
the poor, we should include them for everyone.
Thus, private health insurance paid for by
employers, subsidized student loans, per student expenditures on public education and
even the home mortgage interest deduction
should be figured into the distribution of
resources, making it even more unequal than it
is now. (But, of course, that would not affect
absolute poverty measures.)
More recent criticisms have sought to change
the Orshansky threshold by de-emphasizing its
emphasis on food expenditures as the basis of
what are considered necessities. Specifically,
some researchers argue that the market basket
of necessities has expanded since the early
1960s to include such items as indoor plumbing (which many of the rural poor did not
enjoy in the 1950s) and telephones. Today is
television a necessity? Working heat and air
conditioning? How about a computer? With
these concerns in mind, many analysts have
argued that it is impossible to adjust the
poverty threshold over long periods of time
using the inflation rate (Consumer Price
Index) but that the poverty measure must
be reformulated from scratch every so often
since what is a necessity changes from period
to period, from society to society (Hobjin,
2002).

Calhoun-3267-19.qxd

3/29/2005

2:30 PM

Page 329

THE CONCEPTUALIZATION AND STUDY OF THE POOR

The US poverty threshold is further criticized because it does not take into account
regional variation in the cost of living.5 Living
on 8000 dollars in Mississippi is a lot different
than trying to survive on that same income
in New York City. Finally, there is the issue of
assets and debts. Poverty is measured with
respect to income alone. But income only tells
part of the financial story for most American
families. There is also significant variation in
family wealth levels. (Family wealth also
known as assets or net worth is calculated as
total saleable assets minus outstanding debts
(at the family/household level); insurance,
annuities, received or anticipated social security and other non-fungible assets are not
included in this measure.) This variation in net
worth over and above income levels means
that being poor can be a very different economic experience for families with the same
income levels. This issue is particularly salient
to the study of race, poverty and life chances
in America. Currently, the median African
American family owns about one-eighth the
net worth that the median white family does
(Wolff and Leone, 2002). This difference is not
explained by income or other demographic
characteristics (Oliver and Shapiro, 1995). For
example, among families who earned less than
15,000 dollars per year in 1994, the median net
worth for white families was 10,000 dollars;
the corresponding figure for African American
families was zero. More than half the black
population in this income bracket has no net
worth or is in net debt. Meanwhile, being poor
and white typically means living with a 10,000dollar asset cushion. Income-based measures
obscure this difference. There are many potential ways to integrate income and wealth into a
poverty measure (such as annuitizing wealth
levels and adding them to annual income).
However, policy-makers stick to the traditional
income-based poverty measure; given that
many federal funds are allocated based on the
proportion of a states population that is poor,
the Orshansky line has a political inertia that
is difficult to alter. Likewise, researchers have
only just begun to take into account assets in
explaining the impact of poverty on life chances
largely because good measures of family

329

wealth have not been available until fairly

recently (as of 1984).6
All this criticism leads to the question of
whether an adequate absolute measure of
poverty is ever possible to arrive at. If we
strictly define necessities as those goods without which we cannot survive, then by definition there are no poor among the living.
Recognizing this, most scholars define necessities as what is required to live with dignity. Of
course, if what is necessary to live with dignity
in a given society is socially defined, then is not
every measure of poverty a relative measure?
In other words, if what is a necessity is what
most of us have then there will always be
people who do not have those things in any
market-based economy. In other words, the
poor will always be with us but to greater or
lesser degrees depending on how unequally
income and wealth is distributed. This is one of
the ways that wealth creates poverty by
ratcheting up the social definition of necessity.
Theorists who are of the view that all poverty
is relational have argued for the implementation of relative measures of poverty, most frequently operationalized by considering anyone
with less than one-half the median income to
be poor alternatively, researchers use a cut-off
of 40 per cent of the median income (Fuchs,
1967; see also Rainwater, 1974). This sort of
measure has become standard in the literature
on international comparisons of poverty rate,
since it provides an obvious yardstick that is
commensurable across nations. However, it
really measures income inequality at the bottom half of the distribution.
When using a poverty line set at 40 per cent
of the median income of a given country, a
comparison of poverty rates among developed
nations reveals that the United States is indeed
a laggard with respect to the rest of the developed world. Just less than 11 per cent (10.7) of
the US population enjoys incomes less than
40 per cent of the median. The next closest
country is Australia, with a rate of 7 per cent,
Canada at 6.6 per cent and the United
Kingdom at 5.7 per cent (Smeeding et al., 2000:
Table 2). (If we use the US poverty line and
compare countries, we find that Australia
and the United Kingdom have higher poverty

Calhoun-3267-19.qxd

330

3/29/2005

2:30 PM

Page 330

THE SAGE HANDBOOK OF SOCIOLOGY

rates 17.6 and 15.7 per cent, respectively, as

compared to the 13.6 per cent figure for the US
in the mid-1990s; Smeeding et al., 2000: Table 1).
What is striking is that all of the countries that
are the worst when it comes to relative income
distributions are of Anglo origin following
cultural lines rather than geographical ones.
Once we dip into the middle of the pack, we
find that there is no obvious pattern to which
countries outrank others with respect to their
poverty rates on either absolute or relative
measures. For instance, in absolute rates, what
Gsta Esping-Anderson (1990) calls corporatist welfare state regimes do worse than
social-democratic (or universalist) ones. Using
the US standard, France has a poverty rate of
9.9 per cent and Germany 7.3 per cent. The
Scandanavian countries do better, with
Sweden at 6.3 per cent, Finland 4.8 per cent
and Norway at 4.3 per cent. But when we
switch to the relative rates, Sweden at 4.6 per cent
comes out worse than Germany (4.2) or
France (3.2). It should be clear that what
income is counted, what conversion rate is
used (PPP or exchange rate) and a host of
other issues cloud international comparisons,
making finely graduated rankings almost
meaningless. More important are overall patterns, such as the Anglo countries doing the
worst of all.7

THE EXPERIENCE OF POVERTY:

OUTPUT MEASURES
A newer line of argument coming out of
British and Irish scholarship focuses on outputs rather than inputs. Traditional measures
classify as poor those who do not have the economic ability to meet basic needs within the
behavioral (that is to say, budgeting) expectations of the community. Direct measures of
poverty identify individuals or families whose
actual consumption levels do not meet such
basic needs.8 This distinction carries important implications for the way in which the poor
are conceptualized and treated. In some senses,
the minimum resource conception treats the
poor with more respect than the minimum

standard of living measure since it merely sets

a resource level that is perceived as adequate
and does not make any judgments or assumptions as to how the poor will spend these
resources. A standard of living measure distrusts the poor in a certain way since it implies
that given a certain amount of resources, the
poor may not be efficiently maximizing the
acquisition of necessities in the household ( la
Monaghan). It is also assuming a singular definition of what constitutes a decent standard of
living when groups such as the poor may
differ in what they value.
On the other hand, the minimum standard
of living threshold may better measure the
reality of life for the poor and is not subject to
the vagaries of price and income fluctuations
or differential household needs (such as the
comparison between an elderly couple and a
single mother with a young child illustrates).9
Some researchers have attempted to resolve
this tension between the alternative conceptions of poverty. Stephen Ringen (1988), for
example, advocates the use of a combination of
income thresholds and a measure of deprivation to measure poverty. The concept is
appealing in that people with low incomes but
who are not feeling the pinch on account of
generous in kind benefits, asset wealth, etc.
would not be counted as poor. Likewise, those
who were misers that is, those who experienced deprivation despite being able to afford
to satisfy their needs would not be counted
among the poor either. Only if a household
met the dual criteria of low income and
enforced lack of socially defined necessities
would Ringen consider them poor.
In this vein, many scholars have worked to
develop an adequate measure of deprivation.
The first step is to define a basket of necessities.
For example, one study surveyed residents
of Ireland to come up with a list of socially
defined necessities (including, for example, a
telephone, two pairs of strong shoes, a color
television and a dry dwelling) (Callan et al.,
1993). The poor are those who are low-income
and who lack a certain number of the 20 necessities.10 While this move towards a deprivation
income combination would seem to eliminate
many of the practical problems with input-only

Calhoun-3267-19.qxd

3/29/2005

2:30 PM

Page 331

THE CONCEPTUALIZATION AND STUDY OF THE POOR

measures of poverty, some theorists want to

push further toward a universal set of necessities. These would be constant over time and
place and as such would facilitate comparisons across very diverse cultures and epochs.
The challenge inherent in designing such a
poverty measure is no less than defining
universal goods.
An early such attempt to operationalize a
cross-society, comparable universalistic outputs measure of poverty moving away from
material goods was the Physical Quality of
Life Index (PQLI). Using three basic indicators
of the quality of life, in 1979, Morris David
Morris (1979) constructed a scale for use in
measuring the condition of the worlds populations. The components of this scale are:
infant mortality, life expectancy at age 12
months and basic literacy. Morris admits that
this measure does not capture intangible goods
such as freedom or justice; however, Morris suggests that these three are as close as one can arrive
to a universal that is, non-ethnocentric criterion for well-being. (Human rights, for example, can be defined very differently in one
context than another; and freedom is always a
matter of negotiation between yours and
mine.) Morris purports that all peoples want
their children to live (that is, lower infant mortality) and that all peoples want to live long
themselves (that is, life expectancy). Further,
he claims, even if the desire for literacy per se is
not widely shared literacy could serve as a
surrogate for (although it does not guarantee)
individual capacity for effective social participation (1979: 3). From these three indicators,
he constructs a scale to measure over-time and
cross-sectional differences in the quality of life
of the worlds people.11 The result is a universal
index that allows for comparisons across various societies, sub-populations and cohorts
without running into major incomparability
issues.12 The most obvious drawback is the fact
that by definition individuals or family
units do not have a score on the PQLI, only
populations do.
Nobel Prize winner Amaryta Sen is among
those who would go further than Morris. He
shares with Morris the desire to push for a measure that is less solely reliant on the distribution

331

of material goods, but Sen seeks to expand the

relevant outcomes well beyond those of the
PQLI. The leitmotif that organizes the themes
of poverty and deprivation for Sen is the question of freedom that is, freedom to reach our
full potential. Deprivation, in Sens schema,
occurs when we are prevented from reaching
our full human capabilities. The task at hand
for Sen is to define a basket of human capabilities and then investigate what forces enable
or limit those (e.g. Sen, 1995, 1999, 2000). The
material distribution of resources is a main
but not sole element of the distribution of
freedom to develop our human capabilities.
Wealth makes us free from, say, having to
work, having to stay in one place, having to sell
our dearest possessions and so on but there
are other dimensions of freedom as well.
Freedom, obviously, has a political dimension.
Wealth is less valuable if our voice and our
actions are constrained by a totalitarian state.
Freedom has basic health dimensions as well. If
we do not live long, we enjoy our economic
assets less. In short, you have to be alive to be
free and to develop your human capabilities.
You also have to be relatively absent of crippling disease.
For Sen, financial or material limitations are
but one way that capabilities can be deprived.
That is, market-based distribution mechanisms (economies) represent one allocation
avenue among many. Poverty measures that
examine only material inputs or outputs may
neglect resource inequity based in the political
sphere, for example, or to take another case,
within the household (particularly with
respect to gender inequities) (e.g. Brannen and
Wilson, 1987; Glendinning and Millar, 1987;
Jenkins, 1991). For instance, he claims that
there has never been a massive famine in a
democratically run nation, since elected
regimes are responsive to the needs of their
populations. Politics matters as much as economics. Or rather, they cannot really be separated from each other.
There is a tension in this discussion between
wanting to develop a robust measure of poverty
that accounts for all sorts of deprivation, on the
one hand, and the fear of losing the analytic focus
on specifically economic sources of deprivation,

Calhoun-3267-19.qxd

332

3/29/2005

2:30 PM

Page 332

THE SAGE HANDBOOK OF SOCIOLOGY

on the other. If we broaden the concept of

poverty too widely, we risk making it conceptually and empirically ineffective by conflating a
variety of types of deprivation, oppression and
domination. For example, is it really worthwhile
to talk about political detainees as part of the
poverty problem? We may want to keep these
forms of capabilities deprivation separate. At the
same time, it is not always easy to parse the analytic distinction. Sen and others have shown us
how these worlds act and react on each other.
This is a tension that will not be resolved here but
should serve as grist for sociologists, economists
and political scientists for some time to come.

THE EFFECTS OF POVERTY ON

CHILDRENS LIFE CHANCES
Ultimately, the value of a measure of resources
(such as a poverty line) is how well it predicts
outcomes we care about. There has been much
research assessing the association between
income in general and poverty in particular
and health status, political participation,
deviant behavior and so on. One area in particular that has been a fruitful focus of research
is with respect to the effects of low income
on the outcomes of offspring. An intergenerational lens is particularly appealing to
researchers for both political and methodological reasons. Politically, children are almost
universally seen as members of the deserving
poor in the moral discourse on poverty (and
welfare). In other words, while it may be debated
whether or not a poor adult has reached this
social position through ascription (assignment
through birth) or achievement (through their
own actions), almost by definition it is the case
that children who are poor find themselves in
this condition through ascriptive forces
beyond their control (that is, into which family
they happen to be born). Thus, by focusing
on the effects of economic resources on children,
the current generation of researchers gets away
from the rhetorical trap of blaming the victim
that plagued (and ultimately dampened) much
research on poverty that took place during
the 1960s.

Second, children offer a potential methodological solution to the problems of reverse

and spurious causation. Namely, if we examine
the relationship between, say, poverty and
health among the elderly, we are haunted by
the question of whether any observed association results from poverty causing (ill) health
status; or health status impacting income;
or some third factor say cognitive ability
affecting both. Children with no earnings of
their own would seem to offer the ideal subjects for examining the effects of poverty on
life outcomes since the poverty in which they
may find themselves is largely not due to their
own choices, abilities and so on.13 However,
while a focus on children may largely solve the
reverse causation issues, by itself such a focus
does not adequately address the unobserved
heterogeneity issue, also known as spurious
effects (more on this later).
The research tradition on poverty and child
outcomes is vast and cannot be done justice in
the brief space here (see, e.g., Hauser, 1994).
That said, this literature is perhaps best
summarized in a cumulative, life-course framework over childhood (see Bronfenbrenner,
1979). Starting with birth, much research has
shown that low income and its covariates such
as low maternal education and minority racial
status lead to a greater risk for delivering a
low birthweight baby, due to both prematurity
and intrauterine growth retardation (Cramer,
1995; Gortmaker, 1979; Starfield et al., 1991;
Stockwell et al., 1995). This higher incidence of
low birth weight among the poor population
partially, but not completely, accounts for the
higher infant mortality rates among this group
(Cramer, 1995; Gortmaker, 1979; Tresserras
et al., 1992). If children survive the first year
of life, those from lower SES families face
increased risks of childhood mortality, primarily due to increased chance of accidental death
(Mare, 1982; Wise et al., 1985).14 Aside from
increased mortality rates, children from poor
families suffer from other developmental risks
as well (Egbuonu and Starfield, 1982; Wise and
Meyers, 1988). For instance, there is an inverse
relationship between child blood lead levels
and SES (Brody et al., 1994; Klerman and Parker,
1990; Mahaffey et al., 1982; Quah et al., 1982).

Calhoun-3267-19.qxd

3/29/2005

2:30 PM

Page 333

THE CONCEPTUALIZATION AND STUDY OF THE POOR

Likewise, Sanders Korenman and Jane Miller

(1997) showed that poor children are more
likely to exhibit low height-for-age (stunting)
or low weight-for-height (wasting), two reliable indicators of nutritional status which, in
turn, predict other health outcomes.15 Others
have shown an effect of poverty on childrens
number of bed days and school absences
(McGaughey et al., 1991), on acute illnesses
(Starfield, 1991) and on chronic conditions
such as asthma (Ernst et al., 1995).
In addition to physical health problems,
children from poor families tend to enjoy
worse mental health and display more behavioral problems than their non-poor counterparts, particularly when poverty is long-term
(Campbell, 1995; McLeod and Shanahan,
1993). Poverty may also affect cognitive development. For example, a number of researchers
have found that income is correlated with child
cognitive indicators such as the Peabody
Individual Achievement Tests and the Peabody
Picture Vocabulary Test (Chase-Lansdale et al.,
1997; Duncan et al., 1994; Korenman et al.,
1995).16 Judith Smith and her colleagues found
that between the ages of 3 and 8, relatively
small increases in income can lead to substantial changes in intellectual skills. A one-unit
increase in the ratio of a familys income to a
familys need was associated with a 33.7 point
increase in measures of verbal and math ability
in this study (Smith et al., 1997). At young
ages, children in poverty are also much more
likely than children who are not in poverty to
exhibit behavioral problems in the forms of
aggression, tantrums, anxiety and moodiness.
At older ages, after entrance into school, children in poverty begin to show further disorders in the forms of learning and attention
problems and school disengagement.17 Of
course, however, these income differences were
not randomly assigned.
Furthermore, much of this research has been
done on US and British populations where we
have already seen relative income inequality to
be at its worst in the developed world so it
is not clear whether these effects would persist
in societies where income inequality is less
pronounced and where more basic services
such as housing and childcare have been

333

decommodified. One way to ask about the impact

of poverty on offspring in a cross-national perspective is to look at intergenerational earnings
elasticities. In a meta-analysis Gary Solon (2002)
finds that here too the United States does among
the worst of all, having the lowest degree of
income mobility across generations; to be fair,
however, most of the data from European countries have only respondents who are young and/
or use only single-year measures of income. For
instance, using the German Socio-Economic
Panel, Kenneth A. Couch and Thomas A. Dunn
(1997) find a fatherson elasticity of a multi-year
income measure of only 0.11.18 (Though, using
later waves of the same dataset, Johannes
Wiegand (1997) puts the German figure at 0.34.)
The next lowest estimates of 0.13 come from
Sweden from Swedish income tax records
(three-year averages for both fathers and sons)
and from analysis of the Finnish census data (a
three-year average for the sons and two-year
average for the fathers) (Solon, 2002).19 The
highest estimate of 0.57 comes from the British
National Child Development Survey, where the
respondents (children) were 33 years of age. (It
used a single-year earnings measure for fathers
and only a predicted earnings measure for the
sons, based on education and social class). The
data for the United States, using multiple year
income measures for both fathers and sons,
is upward of 0.40 (Solon, 2002). In this framework, a low fatherson correlation coefficient
might suggest that poverty is less damaging in
that country; or it might not, since income
health gradients have been found in all countries
(though flatter in nations with less income
inequality).
Explanations for these various income gradients can roughly be divided into three camps.
First, some researchers focus on the material
deprivations that low SES induces, such as poor
nutrition, lack of adequate medical care or
unsafe environments (e.g. Callan et al., 1993;
Mack and Lansley, 1985; McGregor and
Borooah, 1992; Ringen, 1987, 1988). Take the
case of food the most basic necessity of all
(along with water). Studies of severe famine in
the Netherlands during the Second World War
have found that consuming fewer than 1000
calories per day results in dramatic reductions

Calhoun-3267-19.qxd

334

3/29/2005

2:30 PM

Page 334

THE SAGE HANDBOOK OF SOCIOLOGY

in pregnancy weight gain and infant birth size

(Lumey and Van Poppel, 1994; Smith, 1947;
Stein et al., 1975). This association between
caloric intake and health displays a very clear
relationship between access to resources and the
biological condition of health and approaches
a causal relationship since the famine was an
exogenous shock to Dutch society. In a contemporary US context, however, results from nutritional studies yield far weaker results. Studies of
diet in the United States have found that, while
poverty increases reported difficultly in affording food, quality of diet itself does not affect
birthweight (Rogers et al., 1998; Widga and
Lewis, 1999). That said, studies examining the
effects of improved nutrition obtained through
participation in the Supplemental Food Program
for Women, Infants and Children (WIC) have
revealed somewhat larger effects. WIC benefits
have been found to reduce low birthweight rates
by up to 25 per cent and very low birthweight
rates by up to 44 per cent (Avruch and Cackley,
1995). But, at the same time, because WIC provides social services beyond supplemental food,
part of these WIC effects may be the result of
factors independent of nutrition. Severe malnutrition is clearly not healthy as data from the
Dutch famine study suggests but malnutrition
also does not appear to be a widespread mechanism connecting economic impoverishment
with poor outcomes in the contemporary United
States and like societies. However, research in
this tradition has gone beyond basic needs such
as nutrition to show that low-income households do experience other forms of material
deprivation, which may explain part of the
effect of income on child cognitive outcomes
(Mayer, 1997). For instance, some work has
shown that poor children are less likely to have
educational toys or books in the household, and
such items are positively associated with healthy
cognitive development (Duncan et al., 1994;
Smith et al., 1997; Zill, 1988; Zill et al., 1991). It
is hard to imagine, however, that toys and books
explain a very large share of the effect of low
income on children.
A second paradigm, often called the parenting stress hypothesis, sees low income, variable
employment, a lack of cultural resources and a
feeling of inferiority from relative social class

comparisons as exacerbating household stress

levels which, in turn, lead to detrimental
parenting practices such as yelling, shouting
and hitting, which are not conducive to healthy
child development (Conger et al., 1992, 1994;
Elder et al., 1995; Hanson et al., 1997; Hashima
and Amato, 1994; Lempers et al., 1989; McLeod
and Shanahan, 1993; Whitbeck et al., 1991).
Further, care for low-income children generally
involves fewer positive interactions between
the child and the caregiver and less opportunity for play (Howes and Olenick, 1986; Howes
and Stewart, 1987; Phillips et al., 1987).
Research suggests that parents living in poverty
are more likely than parents in better conditions to display punitive behaviors such as
shouting, yelling, slapping and less likely to
display love and warmth through behaviors
like cuddling and hugging (Conger et al., 1992,
1994; Elder et al., 1995). A great deal of evidence has connected such parenting practices
to low IQ scores and to behavioral disorders
(Conger et al., 1994).20
What is notable about these two theories of
the effects of poverty on children is how individualistic and behavioralist they are. Poverty,
it seems, can either cause a family not to have
enough material resources, or it can cause the
parents stress, which in turn leads to bad parenting practices. Either way, the causal arrow
runs from the social condition of deprivation
(either absolute in the first case, or relative and
absolute in the second), through the conditions
of the home and the behavior of the parents
and only then to the child. Parents are where
the buck stops either by not providing the
resources their children need or through bad
parenting practices. The bottom line is that
poverty works through the family environment, so the family is ultimately responsible for
mediating its impact on children. Put another
way, poor heroic parents could blunt the deleterious effects by being savvy enough to provide a stimulating educational environment in
the home on the cheap, or by not letting financial stress get between their children and themselves. The direct impact of relative income on
children mediated by peer groups, community conditions and society-wide institutions
such as the mass media are not possible to

Calhoun-3267-19.qxd

3/29/2005

2:30 PM

Page 335

THE CONCEPTUALIZATION AND STUDY OF THE POOR

consider within this framework. Of course, it is

a lot easier methodologically speaking to
look for mediating pathways within the household, rather than through wider social contexts
(more on this later).
Given this ideological frame, it is not surprising that a third theory asserts that it is
not poverty, lack of non-monetary resources
or relative inequality that is so detrimental to
child development as much as it is the fact that
poor parents differ from higher-income parents (Mayer, 1997). Scholars in the no effect
camp assert that the association between SES
and child developmental outcomes is largely
spurious. They claim that the same parental
characteristics that lead to low income, education and occupational prestige also lead to
detrimental developmental outcomes for offspring. These unmeasured characteristics may
range from parenting styles to aspirations to
genetic endowments. This last paradigm
though generally considered significantly more
conservative than the former two shares most
of the same aspects of the material deprivation and parenting stress hypotheses. Namely,
it is a causal (or rather non-causal) story about
parents. The difference boils down to the fact
that the material deprivation and parenting
stress models optimistically believe that mediating factors can be measured and therefore
manipulated, while the no effect camp is
more sanguine on ever explaining the unexplained variance between poor and non-poor
families on child outcomes. This difference has,
of course, enormous political and policy implications and thus is the focus of the ensuing
section.

SLOUCHING TOWARD CAUSATION

The causal pathways that I outlined above for
the material deprivation and parental stress
hypotheses could be altered without reducing
the feasibility of the paradigms. For instance,
rather than a lack of income leading to poor
parenting practices and such parental characteristics then leading to a childs low educational attainment, it could alternatively be

335

the case that parental characteristics are

leading to low income as well as to a childs low
educational attainment. Let us say a parent is
particularly short-tempered; we could imagine
that this tendency would make it hard for
this parent to keep a job, while also having
negative consequences on his or her childs
development. In this case, it is non-economic
characteristics that are leading to economic
circumstances, not the other way around. We
may be dealing here with a case of reverse causation: because certain non-economic characteristics tend to be accompanied by certain
economic characteristics, it can be difficult
to tell whether income is leading to noneconomic characteristics (like temperament and
parenting techniques) or whether such noneconomic characteristics are leading to income
levels. Of course, the order which these different elements follow essentially determines
what (income or non-economic characteristics) is in a position to determine childrens
outcomes.
This potential role of non-economic characteristics in explaining the association between
poverty and childrens outcomes has been
most thoroughly explored by Susan Mayer
(1997) in her book What Money Cant Buy:
Family Income and Childrens Life Chances. In
this book, Mayer takes several steps to untangle
the effects of parental income from parental
characteristics. To begin with, she compares
the effects of different sources of income on
childrens outcomes. Parents may get money
from several different sources earnings, government transfers, etc. and Mayer assumes
that each of these different sources of income
are associated with parental characteristics to
differing degrees. For instance, earnings and
welfare payments are likely oppositely associated with education. Focusing on the effects
of unearned income on kids outcomes, Mayer
compares the effects of parents welfare receipts
(which are strongly associated with parental
characteristics) to other forms of unearned
income (which are so diverse as to be only
weakly associated with parental characteristics).
If income helps children, a dollar from welfare
should be as valuable as a dollar from other
sources of unearned income. Such a uniform

Calhoun-3267-19.qxd

336

3/29/2005

2:30 PM

Page 336

THE SAGE HANDBOOK OF SOCIOLOGY

effect does not appear to be the case, however.

The effect of other forms of unearned income
(that is, not welfare) is smaller than the effect
of total income. Thus, it seems that parental
characteristics may be significantly bound up
in the effects of income.
To further sort out the effects of income and
parental characteristics, Mayer takes advantage
of the role of temporal ordering in causality.
For a factor to cause an outcome, the factor
must generally occur temporally prior to that
outcome. Any statistical effect of a supposedly
causal factor that is found after the outcome
has already taken place cannot possibly be
playing a causal role in the outcome. Using
such logic, Mayer compares the effects of
parental income before an event, like a teenager
having a baby or dropping out of high school,
with the effect of parental income after the
event. If the effect of income after the event is
sizable, it may be assumed that there are significant underlying factors in this measure. Mayer
does, indeed, find that post-event income is a
strong predictor of childrens outcomes, and
argues that income effects on childrens outcomes
may be acting simply as a proxy for unmeasured
parental traits.
Mayer further tests some of the more popular
theories about income and parental traits first,
by comparing how rich and poor parents spend
their money. Mayer finds that high-income parents tend to spend their excess income on larger
homes, cars and eating out more often all
items and practices which likely have little effect
on childrens outcomes. (However, this is questionable since things like a large house may send
subtle messages to children about values and
status.) On the other hand, the material items
that are widely believed to facilitate child development and improve outcomes, such as books
and visits to a museum, Mayer finds are only
weakly related to income. So, rich and poor children appear about equally likely to have the
amenities that are believed to be important to
outcomes. Mayer suggests this is likely because
these items cost so little that their distribution
depends more on parental tastes than actual
income. Next, Mayer considers the effect of
income on parents psychological well-being,
testing the hypothesis that poverty leads to bad

parenting via stress. Mayer finds very little

support for this hypothesis, though, and documents only a weak relationship between parents income and how they interact with their
children. Thus, it also does not appear that
income appreciably influences childrens outcomes through its influence on parents psychological well-being or their parenting practices.
Mayer is quick to note, however, that her findings are only meaningful once childrens basic
material needs are satisfied. In other words, she
interprets her results to mean that once a certain
income threshold is passed, characteristics of
the parents become more important than anything additional money can buy.
Mayers book has received a great deal of
attention and casts serious doubts on much of
the prior research documenting the importance of income on childrens outcomes. In
fact, if explaining poverty without blaming the
victim was the rallying cry for researchers in
the 1960s and 1970s, accounting for unobserved characteristics of rich and poor folks is
the major challenge to researchers in the first
decade of the twenty-first century.
It should be noted, however, that even in a
work as sophisticated as hers we encounter
potential sources of bias. Specifically, some of
Mayers techniques may bias the effects of
income in the opposite direction from traditional analysis that is, toward no effect of
income per se. For example, when considering
the comparison of different sources of parental
income, we must wonder what is included in
the category of parents other unearned
income. Mayer is talking about the following:
inheritance, profit from investments, gifts and
other windfalls. These sources of income are
associated with very atypical events and, therefore, may be related to other changes such as
death of a relative which may have negative
impacts on children. The one source of income
here that would seem the most pure of other
relevant changes investment income is
really moot for the poor since they get almost
none of their income from this source. This
potential role of wealth also means that if income
is non-linear in its effects (as Mayer herself
argues when saying that her results are only
meaningful when basic needs are satisfied), the

Calhoun-3267-19.qxd

3/29/2005

2:30 PM

Page 337

THE CONCEPTUALIZATION AND STUDY OF THE POOR

income changes that are reflected in the effects

of parents other unearned income are largely
among the already well-off, where they should
matter less anyway.
Additionally, Mayers comparison of the
effects of income before and after an event, as
well as her analyses of parents spending habits
and stress levels, could be interpreted as support
for traditional arguments of economic causation. That is, what Mayer calls spurious effects
of income could be indirect effects of income.
We cannot be certain that parental characteristics independent of income at one point
in time are indeed truly separate from prior
economic conditions. That is, parents noneconomic characteristics may in fact be related
to prior socio-economic conditions that Mayer
is unable to measure. If this is, indeed, the case,
we may encounter patterns of economic causation that simply span a very long period of
time.21
Consider, for instance, explanations of differences between rich and poor children that
focus on the so-called culture of poverty.
Some authors posit that because of relative
position at the bottom of the social hierarchy
poor parents develop norms that are problematic in terms of larger society.22 If generations
of irregular employment and discrimination
result in street skills seeming more valuable
than academic skills, parents will be more
likely to encourage their children to acquire
street skills than to study or stay in school.
A simpler account would alternatively state
that parents act as role models not in behaviors
as much as in delineating the possible and
probable in terms of status attainment. That
is, parents who experience a lack of upward
socio-economic mobility no matter their
values or parenting practices may send an
implicit message that achievement and attainment are impossible. These differences in values
and messages are definitely not economic. Yet,
they are so closely associated with economic
conditions that to consider them apart from
income is foolish. Mayers data do not allow
her to consider such possibilities and, thus, the
divisions that she draws between economic
and non-economic factors may not be entirely
definitive.

337

While Mayer has been assailing traditional

poverty research for confusing correlation
with causation, much of the research community has been moving right ahead and addressing larger contextual spheres of economic
inequality. Earlier I mentioned that much of
the research on the impact of poverty on the
health, well-being and life chances of children
had a parental or household focus with respect
to mediating mechanisms. In other words,
causation (or lack thereof) was presumed to
run through the family unit. Recently, however, some researchers have been arguing that
economic inequality at an aggregate level
such as the neighborhood, state or nation
plays an important role in the well-being of
children and populations more generally. This
line of research has two distinct strands worth
mentioning.
First, a substantial literature on neighborhood effects has largely grown out of William
Julius Wilsons book The Truly Disadvantaged.
In this 1987 book, which was partly a response
to conservative arguments about the existence
of an urban underclass made by pundits such as
journalist Ken Auletta in a 1981 New Yorker article.23 Wilson argues that the mobility of jobs
and wealthy urban residents to suburban communities has led to a situation in which the
urban poor are more socially isolated than they
have ever been. Additional researchers have
come along to document both the exodus of
jobs from urban America and the increased segregation of the urban poor. These facts are generally not contested (e.g. Jargowsky, 1997).
Rather, the question that this book and the
larger underclass debate pose is whether the
greater spatial concentration of the poor has an
effect of its own net of the individual level
circumstances in which poor families find
themselves. Hence the question of whether
neighborhood effects exist.24 This is a growing
literature; however, it is a research tradition that
is plagued by the same kind of unobserved
heterogeneity that Mayer takes note of at the
family level. In some senses, the problem of
selection bias may be even worse at the community level thanks to the very trend that Wilson
identifies: self-selection out of poor urban
neighborhoods. That is, given the substantial

Calhoun-3267-19.qxd

3/29/2005

338

2:30 PM

Page 338

THE SAGE HANDBOOK OF SOCIOLOGY

amount of residential mobility that occurs in

and out of neighborhoods, it is not clear
whether the impact of per cent poor in a neighborhood on child (or adult) outcomes such as
school performance, delinquency and crime is a
result of social contagion and environmental
conditions in poor neighborhoods, or simply
the social sorting process that takes place when
families move (or do not move). In other words,
per cent poor and other such community-level
measures may merely be acting as proxies for
unobserved characteristics of the individual
families who live there and not actually picking up some ecological-level trait. Even semiexperimental assessments of programs like
Moving to Opportunity (MTO) in which
public housing residents were scattered into
neighborhoods of varying socioeconomic
circumstances are plagued by issues of selection
bias since it is somewhat voluntary who participates and since there are major issues of contagion between treatment and control groups.
A second potential pitfall for communitylevel research on poverty relates to aggregation
bias when effects of income at the family level
are non-linear.25 Namely, if the effects of
income are non-linear at the individual or
family level (as they should be), aggregate measures may generate spurious correlations if
they are not properly linked back to family
units. This latter issue is much more tractable
than the former issue of selection bias, as it
only requires researchers to shun aggregate
correlations for multi-level models. However,
current researchers should be aware of this
potential hazard to inference.

CONCLUSIONS
There is much research to do with respect to
poverty in rich countries (and the policies that
affect the poor). This chapter has focused on
two research strands among many: (1) the conceptualization measurement of poverty and
(2) the impact of poverty on the life chances of
offspring. There are major challenges to both
these research traditions. Researchers who wish
to conduct robust research and assessment of

poverty over time or across place must come

up with viable measurements that really get at
what we mean by the term poverty in diverse
settings. Researchers who wish to investigate
claims about the impact of household (or community) poverty on the outcomes of children
or adults must grapple with issues of selection
bias if they are going to be taken seriously in
the scientific and public debates around this
important issue. Whether it be through natural
experiments, instrumental variables, family
and community fixed effects models, or
some other innovative statistical approach,
researchers who wish to make claims about the
effects of poverty must go beyond traditional
regression models to have their claims taken
seriously in a social and political environment where it is presumed that the poor and
not poverty are responsible for their own
reproduction.
This chapter has not done justice to wide
swathes of the sociological research tradition
on poverty. I have also not reviewed the illustrious tradition of community, ethnographic
studies of the poor, extending all the way back
to the Chicago School of the early twentieth
century onward through global ethnographies
of the twenty-first. This kind of qualitative
research does much of the legwork in generating the casual stories to undergird the statistical associations that the quantitative poverty
researchers document. These two intellectual
traditions must be in constant dialogue each
moving toward the other in order to solidify the foundation of our knowledge about
economic inequality and deprivation in rich
countries.

NOTES
1 For the text of the speech see Harpers (August 1990),
p. 22.
2 Such a conception fits very well with Karl Marxs
notion of the physical reproduction of labor.
3 She based this multiplier on results from the
Consumer Expenditure Survey of the mid-1950s, which
estimated that families spent on average 35 per cent of
their household budgets on food.
4 Some scholars have called for replacing food with
housing as the basis for need calculations since housing

Calhoun-3267-19.qxd

3/29/2005

2:30 PM

Page 339

THE CONCEPTUALIZATION AND STUDY OF THE POOR

now makes up the largest proportion of household

budgets. See Ruggles: (1990) for a discussion.
5 There is, indeed, a fair amount of evidence that
suggests that the US federal poverty measure is somewhat
crude. Living technically above the poverty line may not
necessarily imply that one has access to the resources one
needs to live with dignity. Families categorized as nearpoor generally have incomes between 100 and 185 per cent
of the poverty line, yet, despite their incomes, frequently
have trouble making ends meet. Further, because the nearpoor are generally ineligible for many government programs, they may be in even more dire straights than the
officially poor again, despite their higher incomes
when trying to provide food, shelter and medical care. For
example, in many states Medicaid is available only to those
families with incomes below 133 per cent of the poverty
line, leaving women and children with low incomes that
are just above the 133 per cent cut off without access to
health care ( Ku et al., 1999; Stevens, 1974).
At the same time, we can also note problems of heterogeneity even among those who fall below the poverty line.
Evidence suggests that poverty comes in several varieties,
and the single measure that accompanies the question of
basic necessities may simply be unable to capture such
diversity. To begin with, there is significant variation in the
duration of poverty, so that some individuals fall into poverty
because of a temporary spell of economic deprivation
often resulting from divorce or unemployment while
others, particularly minorities, may be poor for longer
periods of time with little upward mobility over the life
course. There is also significant variation in the severity of
poverty. In 1999 7 per cent of children lived in extreme
poverty meaning they lived in families with incomes
below 50 per cent of the poverty line (in 1999, the extreme
poverty line was $6145 for a family of three) (Child
Poverty Fact Sheet, 2001).
While the transitory poor and those above the extreme
poverty line far outnumber the consistently poor and the
extremely poor, this inequality in representation is more
than made up for by the implications of duration and
severity (Duncan and Rodgers, 1988). Those who are
persistently poor and those who are extremely poor are at
significantly higher risk for many adverse outcomes, compared to those who are transiently poor, not extremely
poor, and non-poor. For instance, children who experience
prolonged spells of poverty or severe poverty show larger
deficits in cognitive ability and socio-emotional development than children who only experience less severe
poverty or poverty for a short period of time. (For a discussion of poverty and childrens outcomes see Aber et al.,
1997; Duncan and Brooks-Gunn, 1997; Duncan et al.,
1998; Smith and Dixon, 1995).
6 Making this difference in wealth levels all the more
significant is the fact that Conley (1999) has found that
family (parental) wealth is a strong predictor of teenage
and young adult outcomes ranging from teenage premarital childbearing to educational attainment to welfare
dependency to filial wealth accumulation. In many cases,
when parental wealth is taken into account, blackwhite
differences are eliminated or even flip direction. While

339

parental wealth is just beginning to become taken into

consideration in intergenerational studies, it has been
neglected in the race, Socio-economic status (SES) and
child outcomes literature.
7 For a discussion of this see Iceland (2003).
8 Adrian Atkinson (1987) makes a similar distinction
between measures of a minimum level of resources versus
a minimum standard of living.
9 In fact, in the UK Patrick McGregor and Vani Borooah
(1992) find that two substantially different sets of people
are identified as poor depending on what conception is
used.
10 Their list of 20 is based on the work of Joanna Mack
and Stewart Lansley (1985), who sought to develop a deprivation scale.
11 His scale is much simpler than earlier versions which
were developed by the Organization for Economic
Co-operation and Development (OECD), the United
Nations and the United Nations Research Institute for
Social Development (UNRISD), but which later floundered under the political weight of their own complexity.
12 Even with these relatively straightforward measures,
there is a degree of cultural difference in how they are measured. Take infant mortality, for instance. The World
Health Organization has defined a live birth as a product
of conception that shows signs of life irrespective of its
gestational age. However, there is dramatic variation in the
clinical practices of classification. Most US states use the
WHO definition or something close to it. By contrast, a
birth can be recorded up to 48 hours after the time of
delivery in France. As a result, many infants who die before
registration (when infant mortality rates are highest) may
be recorded as a stillbirth. Likewise, in Japan, infants less
than 22 weeks of gestation or with congenital abnormalities are reported as stillbirths regardless of the presence of
signs of life. These examples are meant to show how difficult it is to come up with even the most basic measures
that will be universal across time, place and culture.
13 I say largely since there is a literature in economics
which examines the impact of children on their parents
income. For an example with respect to child health see,
e.g., Rosenzweig and Wolpin (1988).
14 These increased risks may be compounded by the
more limited access to health services on the part of this
population (Newacheck and Halfon, 1986; Perrin et al.,
1989; St Peter et al., 1992).
15 On the effects of nutritional status see Elo and
Preston, 1992; Martorell and Ho, 1984; Miller et al., 1989.
16 For a general summary see Aber et al., 1997; Duncan
and Brooks-Gunn, 1997.
17 For a review see Aber et al., 1997.
18 Their corresponding figure for the United States
(using the Panel Study of Income Dynamics) is only 0.13.
Solon (2002) comments that these low figures are likely to
result from the unusually young ages of the samples.
19 The Swedish estimate comes from Osterberg (2000);
the Finnish result is from Osterbacka (2001).
20 While the stress paradigm has received a great deal of
attention and a fair amount of empirical support particularly
with respect to health status it has also been criticized for

Calhoun-3267-19.qxd

340

3/29/2005

2:30 PM

Page 340

THE SAGE HANDBOOK OF SOCIOLOGY

detracting attention from the primary factor of these models: social position. Bruce Link has argued that research on
the biological consequences of stress is seen as an exciting new development [and] in general, interest has followed the most recent step in the progression toward disease
outcomes, while concern with the earlier foci has dissipated
(Link and Phelan, 1995). In other words, as new developments have emerged, researchers have paid more attention
to the biological specifics of stress and less attention to the
question of why social position is so strongly related to stress
in the first place. Offering an alternative to the stress paradigm, Link has proposed the fundamental cause hypothesis.
Similar to the stress hypothesis, the fundamental cause
hypothesis suggests that relative social positions (e.g.
income levels) have inherent qualities that may promote or
hinder health and child development. Rather than pointing
to inherent levels of stress, however, Link suggests that social
positions provide people with resources like knowledge,
money, power and prestige that can be used to obtain health
resources: As new risk factors become apparent, people of
higher socioeconomic status are more favorably situated to
know about the risks and have the resources that allow them
to engage in protective efforts to avoid them (Link and
Phelan, 1995: 86). For example, some authors have suggested that around the 1960s, when evidence of the risks of
smoking began to emerge, a new class pattern developed in
these behaviors. There is no evidence that, prior to the
1960s, rates of smoking were higher among lower socioeconomic groups. However, during the 1960s individuals of
higher socioeconomic status were more likely to quit
smoking and current research finds strong socioeconomic
gradients in smoking behavior (Ernster, 1988; Norton et al.,
1988). In other words, wealthier people learned about the
heath risks of smoking more quickly than poorer people
and could then mobilize resources to more effectively
change their behavior. The same dynamic may hold with
respect to parenting. In other words, income may be acting
as a proxy for position in a social hierarchy.
21 Other research uses a technique similar to her before
and after approach called sibling fixed effects models,
which compares family income at various times in siblings
childhoods to determine if and when it matters net of
family characteristics that remain constant (such as genes
and the like). One notable study does indeed find an effect
of income early in childhood using some of the same data
that Mayer uses (the PSID): see, Duncan et al., 1998.
22 For discussion of the literature see Mayer, 1997: 50.
23 Wilson was also responding to Charles Murrays
1984 book, Losing Ground, which argued that the persistent plight of the poor was partly a result of aid to the poor.
In other words, an expanding welfare system, Murray
argued, has caused a rise in economic dependency and a
concomitant change in the culture of urban poverty.
24 Some important articles in this tradition include
Brooks-Gunn et al. (1993); Crane (1991); Garner and
Raudenbush (1991); Sampson and Groves (1989).
25 This criticism has been made most vocally at the literature on income inequality and health. For the view that
income inequality is related to health status of adults and

children see, e.g., A. Deaton (2001); Friscella and Franks

(1997); Kennedy et al. (1996, 1998); Lochner (1999);
Soobader and LeClere (1999); Wilkinson (1992).
For the view that it is a statistical artifact of aggregation
bias see, e.g., Gravelle (1998) and Mellor and Milyo (2001).

REFERENCES
Aber, J., Bennett, N., Conley, D. and Li, J. (1997) The
effects of poverty on child health and cognitive
development, Annual Review of Public Health,
18: 46383.
Atkinson, Adrian B. (1987) On the measurement of
poverty, Econometrica, 55: 74964.
Avruch, S. and Cackley, A. (1995) Savings achieved
by giving WIC benefits to women parentally,
Public Health Reports, 110 (1): 27884.
Brannen, J. and Wilson, G. (eds) (1987) Give and
Take in Families. London: Allen and Unwin.
Brody, D.J., Pirkle, J.L., Kramer, R.A., Flegal, K.M.,
Matte, T.D. et al. (1994) Blood lead levels in the
US population, Journal of the American Medical
Association, 272: 27783.
Bronfenbrenner, U. (1979) The Ecology of Human
Development. Ithaca, NY: Cornell University Press.
Brooks-Gunn, J., Duncan, G.J., Klebanov, P.K. and
Sealand, N. (1993) Do neighborhoods influence
child and adolescent development?, American
Journal of Sociology, 99: 35395.
Brooks-Gunn, J., Duncan, G.J. and Aber, J.L.
(eds) (1997) Neighborhood Poverty: Context and
Consequences for Children, Volume I. New York:
Russell Sage Foundation.
Callan, Tim, Nolan, Brian and Whelan, Christopher T.
(1993) Resources, deprivation and the measurement of poverty, Journal of Social Policy, 22:
14172.
Campbell, S.B. (1995) Behavioral problems in
preschool children: a review of recent research,
Journal of Child Psychology and Psychiatry,
36: 11349.
Chase-Lansdale, P., Lindsay, Gordon, Rachel A.,
Brooks-Gunn, Jeanne and Klebanov, Pamela
(1997) Neighbourhood and family influences on
the intellectual and behavioural competence of
preschool and early school-age children, in Jeanne
Brooks-Gunn, Greg Duncan and J. Lawrence Aber
(eds), Neighbourhood Poverty: Volume I. New York:
Russell Sage Foundation. pp. 79118.
Child Poverty Fact Sheet (June 2001) National
Center for Children in Poverty. Available at http://
cpmcnet.columbia.edu/dept/nccp/ycpf.html.

Calhoun-3267-19.qxd

3/29/2005

2:30 PM

Page 341

THE CONCEPTUALIZATION AND STUDY OF THE POOR

Conger, R.D., Conger, K.J., Elder, G.H., Lorenz, F.O.,

Simons, R.L. and Whitbeck, L.B. (1992) A family
process model of economic hardship and adjustment of early adolescent boys, Child Development,
63: 52641.
Conger, R.D., Ge, X., Elder, G.H., Lorenz, F.O. and
Simons, R.L. (1994) Economic stress, coercive
family process, and developmental problems of
adolescents, Child Development, 65: 54161.
Conley, D. (1999) Being Black, Living in the Red: Race,
Wealth and Social Policy in America. Berkeley/
Los Angeles, CA: University of California Press.
Couch, Kenneth, A. and Dunn, Thomas, A. (1997)
Intergenerational correlations in labor market
status: a comparison of the United States and
Germany, Journal of Human Resources, 32: 21032.
Cramer, J.C. (1995) Racial and ethnic differences in
birth weight: the role of income and financial
assistance, Demography, 32: 23147.
Crane, J. (1991) The epidemic theory of ghettos and
neighborhood effects on dropping out and teenage
childbearing, American Journal of Sociology, 96:
122659.
Deaton, A. (2001) Health, inequality and economic
development, Research Program in Development
Studies Working Paper. Woodrow Wilson School
of Government, Princeton University.
Duncan, G. and Brooks-Gunn, J. (1997) Consequences
of Growing up Poor. New York: Russell Sage
Foundation.
Duncan, G. and Rodgers, W. (1988) Has childrens
poverty become more persistent?, American
Sociological Review, 56: 36175.
Duncan, G.J., Brooks-Gunn, J. and Klebanov, P.K.
(1994) Economic deprivation and early childhood
development, Child Development, 65: 296318.
Duncan, G., Brooks-Gunn, J., Yeung, J. and Smith, J.
(1998) How much does childhood poverty affect
the life chances of children?, American Sociological
Review, 63 (3): 40623.
Egbuonu, L. and Starfield, B. (1992) Child health and
social status, Pediatrics, 69: 5507.
Elder, G.H., Van Nguyen, T. and Caspi, A. (1995)
Linking family hardship to childrens lives, Child
Development, 56: 36175.
Elo, I.T. and Preston, S.H. (1992) Effects of earlylife conditions on adult mortality: a review,
Population Index, 58: 36175.
Ernst, P.K., Joseph, D.L., Locher, U. and Becklake, M.R.
(1995) Socioeconomic status and indicators of
asthma in children, American Journal of Respiratory
Critical Care Management, 152: 5705.
Ernster, V. (1988) Trends in smoking, cancer risk,
and cigarette promotion, Cancer, 62: 170212.

341

Esping-Anderson, G. (1990) The Three Worlds

of Welfare Capitalism. Princeton, NJ: Princeton
University Press.
Friedman, Rose (1965) Poverty: Definitions and Perspective. Washington, DC: American Enterprise
Institute.
Friscella, K. and Franks, P. (1997) Poverty or
income inequality as a predictor of mortality:
longitudinal cohort study, British Medical Journal,
314: 17248.
Fuchs, Victor (1967) Redefining poverty and redistributing income, Public Interest, 8: 8895.
Garner, Catherine L. and Raudenbush, Stephen W.
(1991) Neighborhood effects on educational
attainment: a multilevel analysis, Sociology of
Education, 64: 25162.
Glendinning, C. and Millar, J. (eds) (1987) Women in
Poverty in Britain, (Brighton: Wheatsheaf Books.
Gortmaker, S.L. (1979) Poverty and infant mortality
in the United States, American Sociological Review,
44: 28097.
Gravelle, H. (1998) How much of the relationship
between population mortality and unequal distribution of income is a statistical artifact, British
Medical Journal, 316: 3825.
Haber, Alan (1966) Poverty budgets: how much is
enough?, Poverty and Human Resources Abstracts, 1.
Hanson, T.L., McLanahan, S. and Thomson, E.
(1997) Economic resources, parental practices
and childrens well-being, in Greg, J., Duncan and
Jeanne Brooks-Gunn (eds), Consequences of
Growing Up Poor. New York: Russell Sage Foundation. pp. 190238.
Hashima, P.Y. and Amato, P.R. (1994) Poverty,
social support and parental behavior, Child
Development, 65: 394403.
Hauser, R.M. (1994) Measuring socioeconomic status
in studies of child development, Child Development,
65: 15415.
Hobjin, B. (2002) On both sides of the quality bias
in price indexes. Federal Reserve Bank of
New York Staff Report No. 157.
Howes, C. and Olenick, M. (1986) Family and child
influences on toddlers compliance, Child Development, 26: 292303.
Howes, C. and Stewart, P. (1987) Childs play with
adults, toys, and peers: an examination of family
and child care influences, Developmental Psychology,
23: 42330.
Iceland, John (2003) Poverty in America: A Handbook.
Berkeley, CA: University of California Press.
Jargowsky, Paul (1997) Poverty and Place: Ghettos,
Barrios, and the American City. New York: Russell
Sage Foundation.

Calhoun-3267-19.qxd

342

3/29/2005

2:30 PM

Page 342

THE SAGE HANDBOOK OF SOCIOLOGY

Jenkins, Stephen, P. (1991) Poverty measurement

and the within-household distribution: agenda
for action, Journal of Social Policy, 20: 45783.
Kennedy, B., Kawachi, I. and Prothrow-Stith, D.
(1996) Income distribution and mortality: crosssectional ecological study of the Robin Hood Index
in the United States, British Medical Journal,
312: 10047.
Kennedy, B., Kawachi, I., Glass, I. and ProthrowStith, D. (1998) Income distribution, socioeconomic status, and self-rated health: a US multi-level
analysis, British Medical Journal, 317: 91721.
Klerman, L.V. and Parker, M.B. (1990) Alive and
Well? A Review of Health Policies and Programs
for Poor Young Children. New York, NY: National
Center for Children in Poverty, Columbia University
of Public Health.
Korenman, S. and Miller, J.E. (1997) Effects of longterm poverty on physical health of children in
the National Longitudinal Survey of Youth, in
G.J. Duncan and J. Brooks-Gunn (eds), Consequences of Growing Up Poor. New York: Russell
Sage Foundation. pp. 7099.
Korenman, S., Miller, J.E. and Sjaastad, J.E. (1995)
Long-term poverty and child development in the
United States: results form the NLSY, Children
Youth Services Review, 17: 12755.
Ku, L., Ullman, F. and Almeida, R. (1999) What
counts? Determining Medicaid and CHIP eligibility for children, Urban Institute Discussion
Paper No. 9905. Washington, DC: The Urban
Institute Press.
Lempers, J.D., Clark-Lempers, D. and Simons, R.L.
(1989) Economic hardship, parenting and distress
in adolescence, Child Development, 60: 2539.
Link, B. and Phelan, J. (1995) Social conditions as
fundamental causes of disease, Journal of Health
and Social Behavior, Extra Issue: 8094.
Lochner, K. (1999) State inequality and individual
mortality risk: a prospective multi-level study.
PhD Dissertation, Harvard University.
Lumey, L. and Van Poppel, F. (1994) The Dutch
famine of 19441945: mortality and morbidity in
past and present generations, Social Historical
Medicine, 7: 22946.
Mack, Joanna and Lansley, Stewart (1985) Poor
Britain. London: George Allen and Unwin.
Mahaffey, K.R., Annest, J.L. and Roberts, J. (1982)
National estimates of blood lead levels: United
States, 19761980: Association with selected demographic and socio-economic factors, New England
Journal of Medicine, 307 (1): 5739.
Mare, R. (1982) Socio-economic status and childhood mortality, American Journal of Public
Health, 72: 53947.

Martorell, R. and Ho, T.J. (1984) Malnutrition,

morbidity, and mortality, Population and Development Review. Child Survival: Strategies for Research,
10: 4968.
Mayer, Susan (1997) What Money Cant Buy: Family
Income and Childrens Life Chances. Cambridge,
MA: Harvard University Press.
McGaughey P.J., Starfield, B., Alexander, C. and
Ensminger, M.E. (1991) Social environment and
vulnerability of low birth weight children: a
social-epidemiological perspective, Pediatrics,
88: 94353.
McGregor, Patrick P.L. and Borooah, Vani K. (1992)
Is Low Income or Low Expenditure a Better
Indicator of Whether or Not a Household is
Poor, Journal of Social Policy, 21: 5370.
McLeod, J.D. and Shanahan, M.J. (1993) Poverty,
parenting and childrens mental health, American
Sociological Review, 58: 35166.
Mellor, Jennifer M. and Milyo, Jeffrey (2001)
Reexamining the evidence of an ecological
association between income inequality and health,
Journal of Health Politics Policy and Law, 26:
487522.
Miller, C.A., Fine, A. and Adams-Taylor, S. (1989)
Monitoring Childrens Health: Key Indicators, 2nd
edn. Washington, DC: American Public Health
Association.
Monaghan, Thomas (1990) Harpers, August p. 22.
Morris, Morris David (1979) Measuring the Condition
of the Worlds Poor. New York: Pergamon Press.
Murray, Charles (1984) Losing Ground: American
Social Policy, 19501980. New York: Basic Books.
Newacheck, P.W. and Halfon, N. (1986) Access to
ambulatory care services for economically disadvantaged children, Pediatrics, 78: 81319.
Norton, T., Kenneth, E., Juliette, K. and Remington, P.
(1988) Smoking by blacks and whites: socioeconomic and demographic differences, American
Journal of Public Health, 78: 11879.
Oliver, Melvin and Shapiro, Thomas (1995) Black
Wealth, White Wealth: A New Perspective on Racial
Inequality. London: Routledge.
Orshansky, Mollie (1963) Children of the poor,
Social Security Bulletin, US Department of Labor.
Osterbacka, Eva (2001) Family background and economic status in Finland, Scandanavian Journal of
Economics, 103: 46784.
Osterberg, Torun (2000) Intergenerational income
mobility in Sweden: What do tax-data show,
Review of Income and Wealth, 46: 42136.
Perrin, J.M., Homer, C.J. and Berwick, D.M. et al.
(1989) Variations in rates of hospitalization of
children in three urban communities, New England
Journal of Medicine, 320: 11837.

Calhoun-3267-19.qxd

3/29/2005

2:30 PM

Page 343

THE CONCEPTUALIZATION AND STUDY OF THE POOR

Phillips, D., McCartney, K. and Scarr, S. (1987) Child

care quality and childrens social development,
Developmental Psychology, 23: 53743.
Quah, R., Stark, A. and Meigs, J.W. (1982) Childrens
blood levels in New Haven: a population based
information demographic profile, Environmental
Health Perspectives, 5: 12834.
Rainwater, Lee (1974) What Money Buys: Inequality
and the Social Meanings of Income. New York:
Basic Books.
Ringen, Stephen (1987) The Possibility of Politics.
Oxford: Oxford University Press.
Ringen, Stephen (1988) Direct and indirect measures
of poverty, Journal of Social Policy, 17: 35166.
Rogers, I., Emmett, P., Bailer, D. and Golding, J. (1998)
Financial difficulties, smoking habits, composition of diet and birth weight in a population of
pregnant women in the South West of England,
European Journal of Clinical Nutrition, 52 (4): 25160.
Rosenzweig, M.R. and Wolpin, K. (1988) Heterogeneity, intrafamily distribution, and child health,
Journal of Human Resources, 23: 43761.
Ruggles, Patricia (1990) Drawing the Line: Alternative
Poverty Measures and their Implications for Public
Policy. Washington, DC: Urban Institute Press.
Sampson, Robert J. and Byron Groves, W. (1989)
Community structure and crime: testing socialdisorganization theory, American Journal of
Sociology, 94: 774802.
Sen, Amartya (1995) Inequality Reexamined.
Cambridge, MA: Harvard University Press.
Sen, Amartya (1999) Commodities and Capabilities.
Oxford: Oxford University Press.
Sen, Amartya (2000) Development as Freedom.
New York: Anchor Books.
Smeeding, Timothy, Rainwater, Lee and Burtless, Gary
(2000) United States poverty in a cross-national
context, Luxembourg Income Study Working
Paper, No. 244.
Smith, C. (1947) The effects of wartime starvation
on pregnancy and its products, American Journal
of Obstetric Gynecology, 53: 599608.
Smith, J.R., Brooks-Gunn, J. and Klebanov, P.K. (1997)
Consequences of living in poverty for young
childrens cognitive and verbal ability and early
school achievement, in G.J. Duncan and J. BrooksGunn (eds), Consequences of Growing Up Poor.
New York: Russell Sage Foundation. pp. 13289.
Smith, S. and Dixon, R. (1995) Literacy concepts of
low- and middle-class four-year-olds entering
preschool, Journal of Educational Research, 88:
24353.
Solon, Gary (2002) Cross-country differences in
intergenerational earnings mobility, Journal of
Economic Perspectives, 16 (3): 5966.

343

Soobader, M. and LeClere, F. (1999) Aggregation

and the measurement of income inequality:
effects on morbidity, Social Science and Medicine,
48: 73344.
Starfield, B. (1991) Childhood morbidity: comparisons, clusters, and trends, Pediatrics, 88: 51926.
Starfield, B., Shapiro, S., Weiss, J., Liang, K.-Y., Ra, K.,
Paige, D. and Want, X. (1991) Race, family income,
and low birth weight, American Journal of
Epidemiology, 134: 116774.
Stein, Z., Susser, M., Saegner, G. and Marrrolla, F.,
(1975) Famine and Human Development: The
Dutch Hunger Winter of 19441945. New York:
Oxford University Press.
Stevens, R. (1974) Welfare Medicine in America:
A Case Study of Medicaid. New York: The Free Press.
Stockwell, E.G., Goza, F.W. and Roach, J.L. (1995)
The relationship between socioeconomic status
and infant mortality in a metropolitan aggregate,
19891991, Sociological Forum, 10: 297308.
Stone, Deborah A. (1994) Making the poor count,
The American Prospect, 17: 848.
St Peter, R.F., Newacheck, P.W. and Halfon, N.
(1992) Access to care for poor children: separate
and unequal?, Journal of the American Medical
Association, 267: 27604.
Tresserras, R., Canela, J., Alvarez, J., Sentis, J. and
Salleras, L. (1992) Infant mortality, per capita
income, and adult illiteracy: an ecological approach,
American Journal of Public Health, 82: 4358.
Whitbeck, L.B., Simons, R.L., Conger, R.D.,
Lorenz, F.O., Huck, S. and Elder, G.H. Jr (1991)
Family economic hardship, parental support, and
adolescent self-esteem, Social Psychology Quarterly,
54: 35363.
Widga, A. and Lewis, N. (1999) Defined, in-home
prenatal nutrition intervention for low-income
women, Journal of the American Dietetic
Association, 99 (9): 105862.
Wiegand, Johanness (1997) Intergenerational
earnings mobility in Germany, Unpublished paper
cited in Solon (2002).
Wilkinson, R. (1992) Income distribution and
life expectancy, British Medical Journal, 304:
1658.
Wilson, William Julius (1987) The Truly Disadvantaged: The Underclass and Public Policy.
Chicago: University of Chicago Press.
Wise, P.H., Kotelchuck, M., Wilson, M.L. and Mills, M.
(1985) Racial and socio-economic disparities in
childhood mortality in Boston, New England
Journal of Medicine, 313: 3606.
Wise, P.H. and Meyers, A. (1988) Poverty and child
health, Pediatrics Clinic of North America, 35:
116986.

Calhoun-3267-19.qxd

344

3/29/2005

2:30 PM

Page 344

THE SAGE HANDBOOK OF SOCIOLOGY

Wolff, Edward and Leone, Richard (2002) Top

Heavy: The Increasing Concentration of Wealth in
America and What Can Be Done About It, 2nd
edn. New York: The New Press.
Zill, N. (1998) Behavior, achievement, and health
problems among children in stepfamilies: findings from a national survey of child health, in
E.M. Hetherington and J. Arasteh (eds), Impact

of Divorce, Single Parenting and Stepparenting

on Children. Hillsdale, NJ: Lawrence Erlbaum
Associates.
Zill, N., Moore, K. Smith, E., Stief, T. and Coiro, M.J.
(1991) The Life Circumstances and Development
of Children in Welfare Families: A Profile Based on
National Survey Data. Washington, DC: Child
Trends.