Advances in Applied Sociology
2012. Vol.2, No.3, 229-236
Published Online September 2012 in SciRes (
Copyright © 2012 SciRe s . 229
Racial Differences in Perceived Disorder in
Three Gentrifying Neighborhoods
Daniel Monroe Sullivan1, James D. Bachmeier2
1Department of Sociology, Portland State University, Portland, USA
2Population Research Institute, Pennsy l v an ia State University, University Park, USA
Received May 18th, 2012; revised Jun e 20th, 2012; accepted June 30th, 2012
To what extent do diverse residents living in the same neighborhood perceive problems? Do they have
similar levels of concern regarding drug dealing, graffiti, or litter in the streets? This study uses survey
data, probability sampling, and regression analysis to complement qualitative studies that examine per-
ceived disorder in racially diverse, gentrifying neighborhoods. Findings from 571 residents in three
neighborhoods in Portland, Oregon reveal that there are substantial racial differences: white residents
perceive more disorder—both crimes and incivilities—than do blacks. And, in contrast to what contact
theory would suggest, the racial differences are more pronounced among longer tenured residents than
newcomers. Social class—at least the dimension measured by ownership status—is also important, with
homeowners perceiving more than renters. These findings suggest that neighborhoods, and indeed entire
cities, that have racially and socioeconomically diverse residents need to find ways to create stronger so-
cial bonds and solve what Sampson (2009) calls the “paradox of diversity meets disorder”.
Keywords: Race; Social Class; Neighborhoods; Gentrification; Perceived Disorder
I suspect that much of the gentrification debate is actually a
coded reference to the contestation of blacks and whites for
urban spaceAs is so often the case, the action in gentrifica-
tion probably stems from an interaction between race and class
in the urban environment. Because of this interaction, it is dif-
ficult to comprehend the effect of one without taking into ac-
count the other (Massey, 2002: p. 175).
Scholars from across a range of disciplines and countries
recognize that perceived disorder—a social constructivist per-
spective that people to varying degrees understand visible signs
of crime and incivilities as problematic (Sampson & Rauden-
bush, 2004)—can have a big impact on cities (Body-Gendrot,
2000; Bottoms, 2009; Sampson, 2009; Davis, forthcoming) and
neighborhoods, including neighbor relations and the function-
ing of local organizations (Skogan, 1990; Perkins & Taylor,
1996; Sampson & Raudenbush, 2004). Similar perceptions of
order and disorder among residents can facilitate neighborhood
dynamics (Kefalas, 2003; Low, 2003), but differing perceptions
can strain them.
This strain is exacerbated in gentrifying neighborhoods due
to such polarizing conditions as displacement and social class
differences between poorer and less educated longtime resi-
dents and wealthier and more educated newcomers. When gen-
trifying neighborhoods are also going through racial change
—especially from majority black to majority white—the re-
sulting racial distance, mistrust, and tension makes the situation
worse (Anderson, 1990; Sullivan & Shaw, 2011). Researchers
agree that even though some newcomers are attracted to gen-
trifying neighborhoods in part because of its “edginess” and
diversity (Perez, 2002; Lloyd, 2006; Sampson, 2009) they tend
to perceive more disorder, resulting in significant neighborhood
change. Newcomers who perceive more crime, for example,
tend advocate for more police enforcement (Smith, 1996;
Deener, 2010) and against affordable housing (Fraser, 2004;
Sullivan & Picarsic, 2012); newcomers who perceive more
incivilities may, for example, participate in neighborhood or-
ganizations that advocate for residential and commercial “beau-
tification” projects and complain to the city about properties
that do not meet their standard of orderliness (Maly, 2005).
Despite agreeing that newcomers tend to perceive more dis-
order in gentrifying neighborhoods, qualitative methodologies
of previous studies have not been able to examine thoroughly
the salience of race for understanding differences in perceived
disorder. That is because newcomers differ from longtime resi-
dents in a number of important ways: they are more likely to be
white, own their home, and have more money and education.
Our study of residents in three neighborhoods in Portland,
Oregon complements qualitative studies by using survey data
and regression analysis to examine which characteristics are
most salient for understanding differences in perceived disorder.
The racial composition of our sample further enhances our
analysis: the neighborhoods of study have had a substantial
amount of both black and white residents since the 1950s. This
racial diversity allows us to examine if there are differences in
perceived disorder between longtime white and black residents.
Second, qualitative studies tend to focus on highly visible
actors and organizations. Although providing rich description,
it is uncertain how representative they are of neighborhood
residents, most of whom do not publicly express their opinion
about disorder. Less activist residents who perceive high levels
of disorder, however, can influence their neighborhood in im-
portant ways. They may spend less time in public spaces
(Anderson, 1990, 1999; Meares & Kahan, 1998), socialize only
with like-minded neighbors (Hartigan, 1999), not send their
children to the local public school or play in their neighborhood
streets (Prince, 2002), avoid local stores (Anderson, 1990; Maly,
2005), and complain to the police and city agencies (Pattillo,
2007; Deener, 2010). Our study complements these qualitative
studies by randomly selecting 571 residents to measure the
perceptions of a more representative sample. We measure per-
ceived disorder using nine indicators to measure its criminal
and incivilities dimensions and examine its relationship to race,
as well as social class, ownership status, and length of resi-
Perceived Disorder in Gentrifying
Although it is well documented that newcomers in gentrify-
ing neighborhoods tend to perceive more disorder, it is unclear
how salient race and other resident characteristics are for un-
derstanding these differences. Hartigan’s (1999) qualitative
analysis of the racially and socioeconomically mixed Corktown
neighborhood in Detroit illustrates this point. He documents in
rich detail how middle-class white newcomers perceive more
disorder than do long-time black residents, some of whom are
renters. These newcomers tend to socialize in private spaces
with other newcomers, limit their time spent in neighborhood
public spaces, and participate in a neighborhood organization
that advocates closing a local homeless shelter. It is unclear the
extent to which their heightened level of perceived disorder is
influenced by the fact that they are white (who tend to perceive
blacks as disorderly), homeowners (who have a large and rela-
tively fixed investment, are concerned abo ut their property value,
and are drawn to others who share their interests in historic pres-
ervation), college educated (who have been socialized to view
certain neighborhood conditions as disorderly), or newcomers
(who are not yet used to their new neighbors and are instead
comparing them to their previous middle-class ones).
The Race Hypothesis
The literature on residential segregation and racial attitudes
documents that racial stereotyping continues to have a perva-
sive effect on assessments of neighborhood desirability and on
decisions to move in or out (Ellen, 2000). Each group has con-
trasting views of which racial mix constitutes the most desir-
able neighborhood (Farley et al., 1994; Schuman et al., 1997;
Farley, 2011). Blacks, who have learned from historical ex-
perience that majority-white neighborhoods are hostile and that
majority-black neighborhoods are stigmatized and suffer from
disinvestment, tend to prefer a 50-50 mix. Whites, who have
been socialized in a culture that associates blacks with disorder
and are fearful of rising crime (Anderson, 1990) and declining
property values (Farley et al., 1994), tend to be comfortable in
majority-white neighborhoods (Timberlake, 2000; Farley,
2011). This tendency also extends to white gentrifiers, who are
hesitant to move into predominantly black neighborhoods
(Straight, 2010). In both cases the least desirable neighborhoods
are disliked because the particular racial mix conjures images
of disorder. We expect to find, therefore, that in neighborhoods
like ours that have a racial mix that is more in line with the
preference of blacks, black residents perceive less disorder than
In addition, Pattillo (1999) and Woldoff (2011) argue that
black residents, both those living in poor and middle-class
neighborhoods, tend to be exposed to more disorder than whites,
ranging from crimes like drug dealing, gangs, and violence to
incivilities like trash and loud noise. Straight (2010) reaches a
similar conclusion in her analysis of 85 gentrifying neighbor-
hoods: black gentrifiers tend to move into predominantly black
neighborhoods, and ones whose current residents are less stable
and more vulnerable; white gentrifiers, in contrast, tend to
move into white neighborhoods, and ones whose current resi-
dents are more stable and less vulnerable than those encoun-
tered by bla ck gentrifier s.
These findings are explained in large part due to racial seg-
regation and the way in which it concentrates poverty (and
disorder), and the way in which middle-class black neighbor-
hoods are often near poor black ones. The current level of
neighborhood disorder in a racially integrated gentrifying
neighborhood, therefore, would more than likely have to sur-
pass past levels of disorder for it to be perceived by blacks as
problematic. White residents, in contrast, are less likely to have
been exposed to high levels of disorder previously and, thus,
are more likely to perceive current levels of disorder as prob-
The Social Class Hypothesis
Most previous gentrification studies have had a difficult time
disentangling race and social class because of the strong corre-
lation between them (e.g., Kasinitz, 1988; Anderson, 1990;
Smith, 1996; Fraser, 2004; Martin, 2008; Deener, 2010; Lloyd,
2011). A few studies, however, have looked at neighborhoods
undergoing social class changes with no major racial changes
and found that residents with higher socioeconomic status tend
to perceive more disorder. Different dimensions of social class
—ownership status, education, and income—may offer some
insight in explaining the reasons for this finding.
Ownership status. It is reasonable to hypothesize that home-
owners perceive more disorder than do renters since houses are
oftentimes a person’s largest financial investment (Low, 2003)
and they are a relatively fixed investment. If homeowners want
to sell their house, for example because they perceive too much
disorder, it would be cumbersome. They oftentimes invest time
and money to make their house “market ready,” and then they
have to put it on the market, find a buyer, and then buy a house
themselves in another neighborhood. Renters, in contrast, can
simply leave their dwelling (and the neighborhood) when their
contract expir es.
This owner-renter difference helps explain why homeowners
are more likely to exercise what Olson (1965) refers to as
“voice”—i.e., influencing their neighborhood’s dynamics—
rather than “exit”. It also helps explain why homeowners are
more likely to be more sensitive to the reputation of their
neighborhood, since the value of their home is heavily influ-
enced by it (Kasinitz et al., 1998). House values can soar in
neighborhoods that the real estate industry and potential buyers
consider “up and coming,” but values can plummet in neigh-
borhoods considered run-down or crime-ridden (Galster, Tatian
& Smith, 1999; Mele, 2000).
Education. Although positively correlated with homeowner-
ship, education may provide additional explanations for differ-
ences in perceived disorder within gentrifying neighborhoods.
First, since college-educated residents are likely to have been
raised by college-educated parents and socialized by other col-
lege-educated adults, they are more likely to have developed
Copyright © 2012 SciRe s .
tastes for “orderly” behavior (Low, 2003). The most extreme
version of this tendency is residents of gated communities, who
are attracted to a neighborhood where crime is low (due in part
to a private security force and limited public access) and “dis-
orderly” behaviors such as littering and not maintaining yards
are against the rules (Blakely & Snyder, 1997).
Income. Due to residential class segregation, those who are
poor or who were raised in a poor household are more likely to
have spent time living in poor neighborhoods, which tend to
experience higher levels of crime and incivilities (Anderson,
1990, 1999). It seems reasonable to assume, therefore, that
those who have been living near or amongst disorder for a sub-
stantial period of time would be less aware of or bothered by
disorder in their current neighborhood. In addition, those who
are poor tend to have less access to private spaces and therefore
do more activities in the public eye (Duneier, 1999; Pattillo,
2007). It is reasonable to expect that they would be less both-
ered by others engaging in public behavior and, therefore, per-
ceive less disorde r than wealthier residents.
Evidence from previous qualitative studies in gentrifying
neighborhoods supports the social class hypothesis: in gentri-
fying neighborhoods with little racial diversity newcomers tend
to perceive more disorder. In case studies of “black gentrifica-
tion”—i.e., middle-class blacks moving into historically poor,
black neighborhoods—black gentrifiers perceive more disorder.
This is despite the fact that some middle-class black gentrifiers
feel a “deep sense of racial responsibility” (Pattillo, 2007) and a
desire for “racial uplift” (Boyd, 2000). They try to reduce crime
by complaining to the police and joining neighborhood organi-
zations that advocate for more police enforcement (Prince,
2002; Pattillo, 2007). Gentrifiers tend to associate crime with
incivilities and, therefore, make a concerted effort to limit them.
They try to reduce (or at least not increase) poor people in the
neighborhood by limiting homeless shelters, public housing,
and low-income housing (Hartigan, 1999; Fraser, 2004; Pattillo,
2007; Sullivan & Picarsic, 2012), and pressuring landlords to
create more restrictive screening criteria for potential renters
(Pattillo, 2007). They may also be bothered by poor people’s
presence in public spaces, whether it is repairing cars in the
street, barbequing in the park, or “loitering” outside apartment
buildings or bars (Levy & Cybriwsky, 1980; Prince, 2002; Pat-
tillo, 2007).
In addition, they may attempt to change their neighborhood’s
“disorderly” reputation through encouraging positive media
coverage (Kasinitz, 1988), changing the name of the neighbor-
hood (Mele, 2000; Sullivan & Shaw, 2011), and even redraw-
ing neighborhood boundaries to exclude poor, minority resi-
dents who live in public housing (Kasinitz, 1988). Other actions
include promoting “beautification” projects (Prince, 2002),
including forcing local businesses to conform to a middle-class
storefront aesthetic (Kasinitz et al., 1998; Levy & Cybriwsky,
1980; Maly, 2005), courting middle-class potential homebuyers
via neighborhood tours (Kasinitz, 1988), and getting the city to
designate the neighborhood as historic (Kasinitz, 1988; Maly,
Historic preservation designation is particularly effective in
increasing property values (Tournier, 1980) and enforcing a
particular kind of middle- or upper-middle-class sense of aes-
thetics. Brown-Saracino (2004) argues that historic preservation
attracts gentrifiers who embrace historic and architectural quali-
ties and who prioritize historic preservation over “social preser-
vation”—i.e., long-time residents and businesses/organizations,
and their cultural practices.
Years Living in the Neighborhood
Socio-economic differences in perceived disorder could be
exacerbated by how long residents have lived in the neighbor-
hood because old-timers and newcomers may use different
reference points to evaluate the current state of the neighbor-
hood. Residents who have lived in the neighborhood since be-
fore gentrification started, when it suffered from serious crime
and disinvestment and had a substantial amount of vacant lots
and buildings, may perceive less disorder now. More recent
arrivals, in contrast, may have higher expectations since they
are comparing current conditions to those in the neighborhood
from which they moved (or in which they were raised). Simply
put, they may not have lived in their current neighborhood long
enough to notice the changes.
Some scholars also argue that there is a “wave effect” among
gentrifiers (Pattison, 1983; Butler, 2003). First-wave or “risk-
oblivious” gentrifiers are more likely to have a lower socio-
economic status than later-arriving or “risk-averse” ones. Lloyd
(2002) argues that some in-movers during the first stage are
attracted to the “bohemian chic,” which is an attraction to di-
versity, urban grit, and the illicit (see also Perez, 2002; Ley,
2003), because it solidifies their claim to edginess and distin-
guishes them from later-arriving, “inauthentic” Yuppies.
Kerstein (1990) found that later-arriving gentrifiers, in con-
trast, are wealthier and are more likely to have moved from the
suburbs, approve of upscale development, and be attracted to
their house for its architectural and historical value and its po-
tential as a financial investment. And, most important, later-
arriving gentrifiers are more critical of neighborhood services
like garbage collection. These findings suggest that later-arriv-
ing gentrifiers have a higher set of what Low (2003) calls
“norms of middle-class civility” and, therefore, are more likely
to be sensitive to neighborhood disorder.
Contact theory supports this hypothesis, arguing that social
interaction with out-groups improves individuals’ views of
out-group members as they reassess preconceived notions of
the group (Olds, 2012). Although there is still some disagree-
ment about what conditions need to exist for contact theory to
work, many studies have found a positive association between
increased racial contact and positive attitudes toward the
out-group (Siegleman & Welch, 1993; Pettigrew & Troop,
2000). In a similar manner, Anderson (1990) found that the
more time gentrifiers spent living in the neighborhood the more
“streetwise” they were. They became more sophisticated and
nuanced, for example, in assessing whether a situation in the
neighborhood’s public spaces was truly worthy of concern.
On the basis of the above literature review, we will test the
following hypotheses:
1) Whites perceive more neighborhood disorder than do
2) Homeowners perceive more neighborhood disorder than
do renters.
3) Those with a higher socioeconomic status perceive more
neighborhood disorder.
4) The less time a resident has lived in the neighborhood, the
more disorder they perceive.
Copyright © 2012 SciRe s . 231
History of Three Gentrifying Neighborhoods
This paper examines three neighborhoods in Portland, Ore-
gon, which share important similarities. They were predomi-
nantly poor and black since the 1950s, suffering from decades
of disinvestment and housing discrimination (Gibson, 2007).
During the 1980s and early 1990s they experienced high crime,
fueled in part by gang activity and a crack cocaine epidemic.
But during the 1990s these neighborhoods began to gentrify.
They have experienced a decrease in vacant lots, rising housing
prices, and an increase in public and private investment. In
addition, many of the newer residents are white, with mid-
dle-class incomes and college degrees (Burk, 2006). Currently,
all three neighborhoods are diverse in four ways that are related
to gentrification: race, ownership status, social class, and length
of residence in the neighborhood.
The Eliot Neighborhood
Although these three neighborhoods possess some similar
characteri stic s, they di ffer in other respects. The Eliot neighbor-
hood began to gentrify first, starting in the early-to mid-1990s.
This neighborhood is populated by historic churches and
turn-of-the-century Victorian houses, many of which have been
recently renovated. Eliot, however, also has a large amount of
institutionalized housing (e.g., drug and alcohol rehabilitation
facilities). The city has recently renovated Eliot’s main com-
mercial thoroughfare with new sidewalks, bus stops, median
strip with trees, and drinking fountains. This avenue contains
several dozen stores, ranging from a Nike outlet store, coffee-
houses and restaurants, to car repair shops and social service
agencies. Although it is not as thoroughly marketed as artistic
or creative as the Alberta neighborhood, some of its gentrifiers
are part of the “creative class” and express a desire to live
amongst diversity (Florida, 2002).
The Alberta Neighborhood
Gentrification started later in the Alberta neighborhood, from
the mid- to late-nineties, but the rate of change has been faster.
Many businesses on the main commercial strip, as well as some
residents, are making a concerted effort to turn it i nto t he “Al berta
Arts” district (Sullivan & Shaw, 2011). Artists, art galleries, a
monthly art walk, and pu blic art, alon g with ente rprisi ng real esta -
te agents, have cultivated a ne w identi ty, one that has a b ohemian,
“alternative” edge to it. The main commercial street has a food
co-op, a vegan resta ura nt, a numb er of i nde penden t cof fee houses,
yoga studios, and a core group of local business owners a nd resi-
dents who are vehemently against chain stores (Lopas, 2003).
The Concordia Neighborhood
The Concordia neighborhood is a less clear example of gen-
trification. However, it has experienced several changes that
indicate that it is gentrifying. Many of the houses and their
landscapes have been renovated. More white, middle-class
residents have moved in, and services have emerged to serve
them, including an upscale organic grocery store, pharmacy,
coffee shops, and restaurants.
The Concordia neighborhood has several features that would
suggest that its residents would perceive less disorder than
those living in Eliot and Alberta. Its housing stock has always
had the largest share of homeowners, the lowest poverty rate,
the lowest percentage black (whites tend to associate blacks
with disorder), and the lowest crime rate.
Data and Methods
Data come from a face-to-face survey of 571 residents in
three neighborhoods in Portland, Oregon. The sampling frame
consists of housing units that were randomly selected. We sur-
veyed English-speaking adults, but excluded those in institu-
tionally sponsored housing such as drug and alcohol rehabilita-
tion houses since their stay is often temporary and not of their
own choosing. To have a representative sample, surveyors re-
turned many times (and at different times of the day, and on
different days of the week) in an attempt to secure a survey. We
also sent postcards that described the survey to selected house-
holds a few days before going to their door. Given that the ma-
jority of residents self-identify as black or white, non-Hispanic,
we exclude those in other racial and ethnic categories.
The dependent variable is perceived neighborhood disorder,
which consists of five items related to criminal disorder (drug
dealing, assaults and muggings, burglaries and thefts, vandal-
ism and graffiti, and organized gangs) and four items related to
incivilities (loud music, other loud noises, trash in the streets,
and not maintaining lawns or property). Each of the nine items
had three response categories (serious problem, minor problem,
or no problem), resulting in an additive index score ranging
from 0 to 18 (Cronbach’s alpha = 0.794).
Race is the main independent variable and is measured using
a binary that is scored 1 for white, non-Hispanic and 0 for black.
Three other independent variables relate directly to gentrifica-
tion. Socioeconomic status is measured two ways: ownership
status is a binary that is scored 1 for homeowner and 0 for
renter and education is a binary that is scored 1 for college
degree and 0 for less than college degree. Due to the face-to-
face nature of the surveying, we felt that income—another im-
portant dimension of social class—was too personal to ask,
especially given the wide income range of residents (including
some with very low incomes). Length of time living in the
neighborhood is measured in years. Bi-variate analysis indi-
cates that those who have lived in the neighborhood for about
five to ten years perceive more disorder than newcomers and
old-timers. Therefore, we also include the square of years living
in the neighborhood to model this curvilinear relationship.
We also include a number of control variables. Age is meas-
ured in years. Some previous research has found that the elderly
have a greater fear of crime, which is related to perceived dis-
order (Skogan & Maxfield, 1981; Baba & Austin, 1989). Sex is
a binary scored 1 for male and 0 for female. Some findings
indicate that women have a greater fear of crime, again related
to perceived disorder (Skogan & Maxfield, 1981; Perkins &
Taylor, 1996; Austin, Furr, & Spine, 2002). Child at home is a
binary scored 1 for one or more children living at home and 0
for none. One study found parents with children have a greater
fear of crime (Ross & Mirowsky, 2001). The two neighbor-
hood-level variables are binary: Eliot and Alberta; Concordia is
the contrast category.
To examine the relationship between race and perceived dis-
order, as well as ownership status, education, years living in the
neighborhood, and control variables, we use ordinary least
squares regression. The regression has four models: Model 1
just examines the effect of race; Model 2 includes two dimen-
sions of social class: ownership status and education; Model 3
Copyright © 2012 SciRe s .
Copyright © 2012 SciRe s . 233
adds years living in the neighborhood; and Model 4 contains
the control variables and is the full model.
Descriptive Statistics
Table 1 illustrates the descriptive statistics, including for the
whole sample and stratified by race. As is typical of gentrifying
neighborhoods that are going through racial change, whites are
more likely to have a college degree (63.2% v. 18.3%) and own
their home (66.1% v. 48.9%). Black residents have on average
lived in the neighborhood longer (12.8 years v. 8.1) and are
older (45.1 years old v. 40.5).
The overall sample perceives both dimensions of disorder
—crimes and incivilities—as being similarly problematic in
their neighborhoods. About half the respondents, for example,
cite the crimes of burglaries/thefts and drug dealing as prob-
lematic; they are less than half as likely to report organized
gangs and assaults and muggings as such. Similarly, about half
cite the incivilities of trash in the streets and loud noises (other
than music) as problematic.
The Salience of Race
Table 1 also presents the level and type of perceived disor-
der stratified by race.
Supporting our hypothesis at the bi-variate level, whites
overall have a higher perceived disorder index score than
blacks (4.5 v. 3.3). This trend holds true for both perceptions of
criminal disorder and incivilities. Whites, in fact, perceive more
disorder regarding eight of the nine individual disorder items
(only “organized gangs” is not statistically significant).
Given that race is associated with three other important as-
pects of residents living in gentrifying neighborhoods that are
undergoing racial change—whites are more likely to be col-
lege-educated, newcomers, and homeowners—we use regres-
sion analysis to see if race has an independent and nonspurious
effect on perceived disorder. Results from Table 2 further sup-
port the race hypothesis, showing that across all four regression
models blacks are significantly less likely than whites to per-
ceive disorder, even after controlling for two dimensions of
social class and years living in the neighborhood.
The Social Class Hypothesis: Ownership, Not
In addition to race, there is some support for the social class
hypothesis. When examining the two social class variables
simultaneously, Model 4 of Table 2 illustrates that ownership
status is more salient than education in predicting perceived
Table 1.
Descriptive statistics of variables, stratified by race.
Overall Mean Blacks Whites
Dependent Variables
Perceive d Disorder Indexa 4.23b 3.32 4.50
Dimension 1: Percepti ons of Crimec
Burglaries/thefts 50.7%b 35.4% 55.4%
Drug dealin g 49.9%b 41.7% 52.4%
Vandalism/graffiti 44.4%b 30.8% 48.4%
Organized gangs 25.7% 26.2% 25.6%
Assaults/muggings 20.6%b 14.6% 22.5%
Dimension 2: Perce pti ons of Inc ivi lit ies c
Trash in the streets 52.3%b 36.9% 56.8%
Loud noises (other than music) 45.5%b 33.6% 49.1%
Not maintain in g l awns/pro perty 38.5%b 30.5% 40.9%
Loud music 36.0%b 29.2% 38.0%
Independent Variables
College De gree 52.8%b 18.3% 63.2%
Homeowner 62.2%b 48.9% 66.1%
Years Li v ing i n Neighbo rhood 8.07b 12.77 6.67
Age 40.50b 45.06 39.15
Gender (Male) 45.1% 43.8% 45.5%
Child Living in Home 28.0%b 48.1% 22.0%
Eliot Neighborhood 32.6% 39.7% 30.5%
Alberta Neighborhood 37.8% 35.1% 38.6%
Concordia Neighborhood 29.6% 25.2% 30.9%
N 571 131 440
aThe Perceived Disorder Index ranges from 0 - 18. bChi-square test of proportional differences between blacks and whites are significant at the 0.05 level.
cResidents perceived them as a minor or serious pr oblem.
Table 2.
OLS regression estimates of perceived disorder.a
Model 1 Model 2 Model 3 Model 4
Race (Blacks) –1.18*** –1.01** –.98** –.90**
College De gree .26 .21 .15
Homeowner .39 .43 .76*
Years in Neighborhoo d .04 .06
Years in Neighborhoo d (s quared) –.01 –.01*
Age –.01
Gender (Male) .16
Child Living in Home –.50
Eliot Neighborhoodb 1.49***
Alberta Ne ighborhoodb 1.48***
Constant 4.50*** 4.09*** 4.01*** 2.78***
Adjusted R 2 .02 .02 .03 .07
*p < .05; **p < .01; ***p < .001; aPerceived disorder is measured by an index (0 - 18); bThe Concor di a ne igh borhood is the co nt ra st ca te g ory.
disorder. Even after controlling for race and education, home-
owners are still more likely to perceive neighborhood problems
than are renters. This supports the argument that homeowners
are more sensitive to neighborhood problems, perhaps due to
the large and relatively fixed nature of their house investment
and their inability to exit easily from the neighborhood. The
other indicator of social class—education—is not statistically
significant once ownership status and race are taken into ac-
count. These findings suggest that the social class indicator
most directly tied to residents’ financial well-being—home-
ownership—more strongly correlates to perceived disorder.
Years Living in the Neighborhood Matters but…
Our “years living in the neighborhood” hypothesis is only
partially supported. We hypothesized that longer tenured resi-
dents would perceive less disorder since there is now less crime,
fewer vacant lots and buildings, and other tangible signs of
disorder than in the past. Many of the newer arriving residents,
most of whom have more money than longer-time residents,
have probably moved in from more “orderly” neighborhoods
and, therefore, would perceive more disorder in their current
neighborhood 1) because they would not have been around long
enough to witness the decline in crime, vacant lots/buildings,
etc. and 2) because their point of reference is more likely to be
a more “orderly” neighborhood. After conducting a bi-variate
analysis with perceived disorder, however, we realized that our
respondents who lived in the neighborhood for about five to ten
years perceive the most disorder, and newcomers and long-time
residents perceived less. (That is why we added a squared ver-
sion of “years living in the neighborhood” variable to the re-
gression models.)
Perhaps newer residents are not as invested in or knowl-
edgeable about neighborhood conditions as those who have
lived in the neighborhood longer. Indeed, further analyses (not
reported) find that newcomers are less likely to have family or
friends living in the neighborhood, and they are also less likely
to be involved in neighborhood organizations, like churches
and neighborhood associations.
In Figure 1 we examine whether racial differences in per-
ceived disorder fade as residents spend more time living in the
neighborhood. That is, as black and white residents reside to-
gether over time, do they arrive, as contact theory suggests, to a
shared understanding of how they perceive their community,
even if such sentiments are not shared when they first become
neighbors? None of the results suggest that they do. For each of
the five “years living in the neighborhood” categories, whites
perceive more disorder. And the racial gap, rather than getting
smaller over time, gets larger. The smallest racial gap is for
relative newcomers (0 - 5 years in the neighborhood), and then
it grows larger for the next tenured group (5 - 10 years), and
even larger for those living in the neighborhood for more than
ten years.
Model 4 of Table 2 indicates that residents in the Eliot and
Alberta neighborhoods perceive more disorder. One the one
hand, this is not surprising since these two neighborhoods are
gentrifying more rapidly and extensively and they have more
characteristics associated with perceived disorder—e.g., more
poverty, minority residents, crime, and renters—than do Con-
cordia residents. On the other hand, gentrifiers in these two
neighborhoods correspond more closely to the “bohemian” or
“alternative” residents that Perez (2002), Lloyd (2002, 2006),
and Sampson (2009) discuss. Despite their supposed attraction
to diversity and “edginess”, they perceive more disorder than
do those living in the more culturally mainstream Concordia
neighborhood. The control variables of age, gender and whether
they have a child living at home are not significant.
Figure 1.
Perceived disorder, by race an d years living in the neighbor hood.
Copyright © 2012 SciRe s .
This study contributes to our understanding of race relations
in urban neighborhoods by demonstrating the salience of race
when examining perceived disorder in gentrifying neighbor-
hoods. We complement qualitative studies by using a random
sample of 571 residents, which allows us to understand the
perceptions of both activist and non-activist residents. We also
employ regression analysis to examine the influence of race
while statistically controlling for other important and highly
correlated residential characteristics. Finally, we examine
neighborhood types that have been largely ignored: socio-eco-
nomically and racially diverse neighborhoods that have a bal-
ance between newcomers and long-time residents and between
homeowners and renters.
Our findings support the race hypothesis: whites perceive
more disorder than blacks—both crimes and incivilities—after
controlling for ownership status, education, and years living in
the neighborhood. And these racial differences do not decrease
over time, as contact theory would suggest. In fact, we find the
opposite: there are greater racial differences among longer ten-
ured residents than newer ones.
Douglas Massey’s provocative quote that opened this article
regarding the salience of race resonates with our findings. Fu-
ture research should investigate whether the results from this
study are representative of other gentrifying neighborhoods.
Portland’s black population is smaller than it is in other U.S.
cities, and white residents may therefore perceive blacks as less
of a racial threat or source of disorder (Brown & Warner, 1992;
Krueger & Mueller, 2001). Portland, like other western US
cities, also has had less racial segregation historically than cities
in other parts of the US (Massey & Denton, 1993). For these
reasons, we anticipate greater racial differences in perceived
disorder in cities with a larger black population and more racial
segregation. Future research should also examine whether our
findings hold true in cities in other countries and in regards to
other racial and ethnic groups.
The salience of race does not mean that social class is unim-
portant for understanding tension and social isolation in gentri-
fying neighborhoods; previous qualitative studies have demon-
strated clearly that it does. What our findings suggest, however,
is that among the different dimensions of social class, owner-
ship status is more influential than education. In particular,
homeowners perceive more disorder than do renters. We
speculate that homeowners are more sensitive to disorder be-
cause their house is a major, and relatively fixed, investment. It
would be difficult for them to leave the neighborhood if prob-
lems became too severe and the move would probably result in
a financial loss. The lack of statistical significance of education
suggests that the social class indicator most closely associated
with financial considerations—ownership status—is more sali-
ent than the social class indicator that is associated more with
culture. One weakness of the current study that future research
can address is the salience of income in understanding differ-
ences in perceived disorder.
Overall, research has illustrated how perceived disorder leads
to powerful neighborhood consequences: e.g., greater police
presence and scrutiny of poor and minority residents, pressure
to reduce housing for the poor and marginalized, and social
isolation between newcomers and old-timers. Given the impor-
tance of race in explaining differences in perceived disorder,
future research should explore ways to promote what Weisinger
& Salipante (2005) call “racially bridging ties”: relations be-
tween organizations that represent the interests of different
racial groups. Future research should also examine how
neighborhood associations, which in the U.S. are chartered to
represent the interests of all its residents, can serve as places for
diverse residents in gentrifying neighborhoods to come together
to define and solve problems collectively and promote genuine
social integration. And it is essential for researchers and poli-
cymakers to more fully understand and resolve what Sampson
(2009) calls the “paradox of diversity meets disorder”: neighbor-
hoods and cities need to find ways for the creative class, who
are supposedly attracted to diversity (Florida, 2002), to create
meaningful bonds with already existing residents.
Anderson, E. (1990). Streetwise. Chicago, IL: The University of Chi-
cago Press.
Anderson, E. (1999). Code of the street. New York: W. W. Norton and
Aus tin , M. D ., Fu rr, A . L. , & Spine, M. (2002). The effects of neighbor-
hood conditions on perceptions of safety. Journal of Criminal Justice,
30, 417-427. doi:10.1016/S0047-2352(02)00148-4
Baba, Y., & Austin, M. D. (1989). Neighborhood environmental satis-
faction, victimization, and social participation as determinants of
perceived neighborhood safety. Environment and Behavior, 21, 763-
780. doi:10.1177/0013916589216006
Blakely, E. J., & Snyder, M. J. (1997). Fortress America. Washington,
DC: Brookings Institution Press.
Body-Gendrot, S. (2000). The social control of cities? Oxford: Black-
well. doi:10.1002/9780470712757
Bottoms, A. (2009). Disorder, order and control signals. British Journal
of Sociology, 60, 49-55. doi:10.1111/j.1468-4446.2008.01214.x
Boyd, M. (2000). Defensive development. Urban Affairs Review, 43,
751-776. doi:10.1177/1078087407313581
Brown, C. M., & Warner, B. D. (1992). Immigrants, urban politics, and
policing in 1990. Am e r i c an S o c i o l o gical Review, 57, 293-305.
Brown-Saracino, J. (2004). Social preservationists and the quest for
authentic community. City & Community, 3, 135-156.
Burk, B. (2006). A decade of change on Alberta. The Portland Ob-
server, 11 January.
Butler, T. (2003). Living in the bubble. Urban Studies, 40, 2469-2486.
Davis, D. E. (forthcoming). The Giuliani factor. In G. A. Jones (Ed.),
Public sphere and public space in Mexico City, London: Palgrave
Deener, A. (2010). The “black section” of the neighborhood. Ethnog-
raphy, 11, 45-67. doi:10.1177/1466138109346988
Duneier, M. (1999). Sidewalk. New York: Macmillian.
Ellen, I. G. (2000). Sharing America’s neighborhoods. Cambridge, MA:
Harvard University Press.
Farley, R., Steeh, C., Krysan, M., Jackson, T., & Reeves, K. (1994).
Stereotypes and segregation. American Journal of Sociology, 100,
750-780. doi:10.1086/230580
Farley, R. (2011). The waning of American apartheid? Contexts, 10,
36-43. doi:10.1177/1536504211418452
Florida, R. (2002). Rise of the creative class. New York: Bas ic B oo k s .
Fraser, J. C. (2004). Beyond gentrification. Urban Geography, 25, 437-
457. doi:10.2747/0272-3638.25.5.437
Galster, G. C., Tatian, P., & Smith, R. (1999). The impact of neighbors
who use Section 8 certificates on property values. Housing Policy
Debate, 10, 879-917.
Gibson, K. (2007). Bleeding albina. Transforming Anthropology, 15,
3-25. doi:10.1525/tran.2007.15.1.03
Hartigan, J. (1999). Racial situations. Princeton, NJ: Princeton Univer-
sity Press.
Copyright © 2012 SciRe s . 235
Copyright © 2012 SciRe s .
Kasinitz, P. (1988). The gentrification of “Boerum hill”. Qualitative
Sociology, 11, 163-182. doi:10.1007/BF00988953
Kasinitz, P., Bazzi, M., & Doane, R. (1998). Jackson heights, New
York. Cityscape, 4, 161-177.
Kefalas, M. (2003). Working-class heroes. Berkeley: University of
California Press.
Kerstein, R. (1990). Stage models of gentrification. Urban Affairs
Quarterly, 25, 620-639. doi:10.1177/004208169002500406
Krueger, B. S., & Mueller, P. D. (2001). Moderating backlash. State
Politics & Policy Quarterly, 1, 165-179.
Levy, P. R., & Cybriwsky, R. A. (1980). The hidden dimensions of
culture and class. In S. B. Laska, & D. Spain (Eds.), Back to the city
(pp. 138-155). New Yor k: Pergamon.
Ley, D. (2003). Artists, aestheticisation and the field of gentrification.
Urban studies, 40, 2527- 2544.
Lloyd, R. D. (2002). Neo-bohemia. Journal of Urban Affairs, 24, 517-
532. doi:10.1111/1467-9906.00141
Lloyd, R. D. (2006). Neo-bohemia. New Yo rk: Routledge.
Lloyd, R. D. (2011). East Nashv ille skyline. Ethnography, 12, 114-145.
Lopas, M. (2003). Whispers of big chain unsettle Alberta’s independent
spirit. The Oregonian, 7 February.
Low, S. (2003). Behind the gates. New York: Routledge.
Maly, M. (2005). Beyond segregation. Philadelphia: Temple University
Martin, L. (2008). Boredom, drugs, and schools. City & Community, 7,
331-346. doi:10.1111/j.1540-6040.2008.00270.x
Massey, D. S. (2002). Comment on “does gentrification harm the
poor?” Brookings-Wharton Papers in Urban Affairs. Washington,
DC: Brookings Institution.
Massey, D. S., & Denton, N. A. (1993). American apartheid. Cam-
bridge, MA: Harvard University Press.
Meares, T. L., & Kahan, D. M. (1998) Law and (norms of) order in the
inner city. Law and Society Review, 32, 805-838.
Mele, C. (2000). Selling the lower eastside. Minneapolis, MN: Univer-
sity of Minnesota Press.
Olds, C. (2012). What role can propinquity play in the development of
new national allegiances? Advances in Applied Sociology, 2, 7-18.
Olson, M. (1965). The logic of collective action. Cambridge, MA: H ar-
vard University Press.
Pattillo, M. (2007). Black on the block. Chicago, IL: University of
Chicago Press.
Pattillo, M. (1999). Black picket fences. Chicago, IL: University of
Chicago Press.
Pattison, T. (1983). The stages of gentrification. In P. Clay, & R. Hol-
lister (Eds.), Neighborhood policy and planning (pp. 77-92). Lex-
ington, MA: D. C. Heath.
Perez, G. M. (2002). The other “Real World.”. Urban Anthropology, 31,
Perkins, D. D., & Taylor R. B. (1996). Ecological assessment of com-
munity disorder. American Journal of Community Psychology, 24,
63-107. doi:10.1007/BF02511883
Pettigrew, T., & Troop, L. (2000). Does intergroup contact reduce
prejudice? In S. Oskamp (Ed.), Reducing prejudice and discrimina-
tion (pp. 93-114). New Jersey: Lawrence Earlbaum Associates, Inc.
Prince, S. R. (2002). Changing places. Urban Anthropology, 31, 5-35.
Ross, C. E., & Mirowsky, J. (2001). Neighborhood disadvantage and
health. Journal of Health and S o c i a l Behavior, 42, 258-276.
Sampson, R. J. (2009). Analytic approaches to disorder. British Journal
of Sociology, 60, 83-93. doi:10.1111/j.1468-4446.2008.01219.x
Sampson, R. J., & Raudenbush, S. W. (2004). Seeing disorder. Social
Psychology Quarterly, 67, 319-342.
Schuman, H., Steeh, C., Bobo, L., & Krysan, M. (1997). Racial atti-
tudes in America. Cambridge, MA: Harvard University Press.
Sieglman, L., & Welch, S. (1993). The contact hypothesis revisited.
Social Forces, 71, 781-795.
Skogan, W. G. (1990). Disorder and decline. Berkeley, CA: University
of California Press.
Skogan, W. G., & Maxfield, M. G. (1981). Coping with crime. New-
bury Park, CA: Sage Publ i c ations.
Smith, N. (1996). The new urban frontier. New York: Routledge.
Straight, A. J. (2010). Location isn’t everything. Atlanta, GA: Ameri-
can Sociological Association.
Sullivan, D. M., & Picarsic, J. (2012). The subtlety of social exclusion.
Sociology Mind, 2, 153-157. doi:10.4236/sm.2012.22020
Sullivan, D. M., & Shaw S. C. (2011). Race and retail gentrification.
Urban Affairs Review, 47, 403-422.
Timberlake, J. M. (2000). Still life in black and white. Sociological
Inquiry, 70, 420-445. doi:10.1111/j.1475-682X.2000.tb00918.x
Tournier, R. E. (1980). Historic preservation as a force in urban change.
In S. B. Laska, & D. Spain (Eds.), Back to the city. New York: Per-
gamon Press.
Weisinger, J. Y., & Salipante, P. F. (2005). A grounded theory for
building ethnically bridging social capital in voluntary organizations.
Nonprofit and Voluntary Sector Quar terly , 34, 29-55.
Woldoff, R. (2011). White flight/black flight. Ithaca: Cornell University