ArticlePDF Available

Who Is More Dangerous? Comparing the Criminality of Adult Homeless and Domiciled Jail Inmates: A Research Note

DOI:10.1177/0306624X00441006
Authors:
Matt Delisi at Iowa State University
Matt Delisi
Download full-text PDF
Read full-text
Download citation

Abstract and Figures

The criminality of 100 homeless and 100 domiciled jail inmates was compared. Homeless jail inmates were significantly more likely than domiciled jail inmates to be mentally ill, to be arrested for nuisance offenses, to have more extensive criminal histories, and to have prior arrests for use of weapons, drugs, and alcohol. Suggestions for processing homeless offenders are given.
Figures - uploaded by Matt Delisi
Author content
All figure content in this area was uploaded by Matt Delisi
Content may be subject to copyright.
Content uploaded by Matt Delisi
Author content
All content in this area was uploaded by Matt Delisi on Jan 27, 2014
Content may be subject to copyright.
http://ijo.sagepub.com
Comparative Criminology
International Journal of Offender Therapy and
DOI: 10.1177/0306624X00441006
2000; 44; 59 Int J Offender Ther Comp Criminol
Matt DeLisi
Research Note
Who Is More Dangerous? Comparing the Criminality of Adult Homeless and Domiciled Jail Inmates: A
http://ijo.sagepub.com/cgi/content/abstract/44/1/59
The online version of this article can be found at:
Published by:
http://www.sagepublications.com
can be found at:International Journal of Offender Therapy and Comparative Criminology Additional services and information for
http://ijo.sagepub.com/cgi/alerts Email Alerts:
http://ijo.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
InternationalJournal ofOffender Therapy and ComparativeCriminology
Criminalityof Adult Homeless
Who Is More Dangerous?
Comparing the Criminality
of Adult Homeless and Domiciled
Jail Inmates: A Research Note
Matt DeLisi
Abstract: The criminality of 100 homeless and 100 domiciled jail inmates was compared.
Homelessjailinmatesweresignificantlymorelikelythandomiciledjailinmatestobementally
ill, to be arrested for nuisance offenses,to have moreextensive criminal histories,and to have
prior arrests for use of weapons, drugs, and alcohol. Suggestions for processing homeless
offenders are given.
Homeless people in the United States are a source of frequent and varied socio-
logical inquiry. Researchers are ideologically torn over whether the homeless are
worthyor unworthyof public sympathyandsupport (Wright, 1988b).Aclassical
schoolargument(e.g., CesareBeccaria andJeremyBentham)isthatthehomeless
arerational,free-thinkingactorswhohavenoonebutthemselvestoblameforper-
sonal inadequacies such as alcoholism, drug addiction, unemployment, tran-
siency,and mental illness.The homeless, accordingto a classical perspective, are
a reproachablegroup of derelicts. Conversely, apositivist school argumentis that
the homeless are a disparate group of unfortunate, indigent people whose social
conditionisattributabletomacro-societal forcessuch asa changingeconomy,not
personal inadequacies. The homeless, according to a positivist perspective, are
unfortunate victims of social forces.
Investigations of homeless criminality generally occupy four categories: (a)
whether the homeless involvement in the criminal justice system is legitimate or
the result of police harassment (Aulette & Aulette, 1987; Irwin, 1985), (b)
whether a real relationship exists between mental illness and homeless criminal-
ity(Belcher,1988;Benda,1987;Simons,Whitbeck,& Bales,1989; Snow,Baker,
Anderson, & Martin, 1986; Wright, 1988a), (c) whether a relationship exists
between drug and alcohol abuse and homeless criminality (Snow, Baker, &
Anderson, 1989), and (d) whether homelessness itself is criminogenic (McCarthy &
Hagan, 1991, 1992).
International Journal of Offender Therapy and Comparative Criminology, 44(1), 2000 59-69
2000 Sage Publications, Inc.
59
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
PRIOR HOMELESS CRIME RESEARCH
Some researchers (Aulette & Aulette, 1987; Irwin, 1985) contend that the
homelessarevictimsof unnecessarypoliceharassment, whichhelpsexplaintheir
disproportionately high arrest rates. Consider this passage from Aulette and
Aulette (1987):
[Accordingto the police]findinga place to sleepand urinateis trespassing.Waiting
to eat and sell blood are looked upon as loitering. Trying to get cigarettes or a free
busrideis panhandling.Andcarrying aroundone’s belongingsis“squatting” orcar-
rying concealed weapons. In short, staying alive is a crime. (p. 253)
According to these authors, homeless criminality is the outcome of the unjust
criminalization of transient life.
The homeless also have disproportionately high incarceration rates. Irwin
(1985) argues that jails function to manage detached and disreputable transient
social misfits called “rabble.” Rabble are social nuisances who commit petty, not
dangerous crimes. Indeed, a majority of arrests for the homeless are for public
intoxication, theft/shoplifting, violation of municipal ordinances, and burglary
(Snowet al.,1989).Thus,some (Aulette& Aulette,1987;Irwin, 1985)determine
that highcriminality among the homelessis an artifactof unnecessary police har-
assment or susceptibility to arrest, not actual criminal offending.
A common lay perception is that the homeless are a population plagued by
severe mentalillness. Researchersdisagree aboutthe prevalenceofmental illness
among thehomeless. Snowet al. (1986) claimthat the notion ofpervasivemental
illness among the homeless is a myth. In a study of 1,000 homeless adults in
Texas, Snow et al. (1986) offered four reasons for the homeless–mental illness
misconception. First, they contend that too much causal emphasis is afforded to
de-institutionalization as the main cause of the growing homeless phenomenon.
Second,theyarguethattheheightenedvisibilityofhomelesspeoplewhoaremen-
tally ill contributes to the belief that all homeless people are mentally ill. Third,
the medicalization of the homeless is premature. Fourth, studies that found high
rates of mental illness among the homeless are methodologically invalid. The
authorsfoundthatonly10%oftheTexashomelessintheirsamplecouldbeclassi-
fied as having legitimate psychiatric problems. However, Wright (1988a) repli-
catedtheSnow etal.(1986) studyand determinedthat33% ofthe Texashomeless
were mentally unhealthy.
Belcher (1988) developed a typology that ranked the extent of mental illness
among the homeless. Category I is people with chronic mental illness and an
extended period of transiency. Category II is people with minor affective disor-
ders who arestill able to retain employmentdespite being homeless. CategoryIII
is people whose homeless condition is due to personality disorders often aggra-
vated by substance abuse. Category IV is intentionally homeless people with
severe substance abuse problems. Implicit in this typology is the idea that some
60 International Journal of Offender Therapy and Comparative Criminology
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
people are justifiably homeless whereas others are not. Zapf, Roesch, and Hart
(1996)reportedthat10%ofhomelessjail inmatessufferedfromseverementalill-
ness whereas nearly two thirds suffered from some lesser form of mental illness.
However, the only statistically significant difference between homeless and
domiciled jail inmates in their study was that the homeless were more likely to
suffer from negative psychotic symptoms such as emotional withdrawal, motor
retardation, and blunted affect.
Some havefoundhomelessness to be inherently criminogenic.Intests of Gib-
bons’s (1971) situational explanation of crime, McCarthy and Hagan (1991,
1992) found that crime was a recourse for homeless people in that adverse situa-
tional conditions (hunger, lack of shelter, and unemployment) lead to theft of
food, serious theft, burglary, and prostitution. The authors contend that involve-
ment in crime is positively correlated with length of time on the streets. Finally,
Snow et al. (1989) reported several important findings in the epidemiology of
homelesscrime.First,83%ofthehomelessintheirstudyperpetratedPartIIIndex
crimes, which were victimless. Second, 37% of the homeless perpetrated Part I
Indexcrimesinvolvinganonhomelessvictim.Third,lessthan6%ofthehomeless
perpetrated Part I Index crimes involving a homeless victim. Fourth, 56% of the
homeless perpetrated Part I Index property crimes involvingacommercial estab-
lishment as victim. Although a majority of homeless-perpetrated crimes were
relatively benignPartII Indexoffenses,thehomelessposeda dangertononhome-
less people and commercial establishments.
For the current study, homeless jail inmates are defined as offenders without a
physicaldomicilefor morethanayearprior toarrest.Oneyearof homelessnessis
referred to as chronic homelessness or chronic transiency. Of the homeless sam-
ple, 82% (n = 82) reported chronic homelessness for the entirety of their adult
lives, that is, they had never had a domicile. Domiciled jail inmates are arrestees
with a domicileforat leasta yearprior toarrest.Of thedomiciled sample,4% (n=4)
had ever experienced a period of homelessness. None ofthe100domiciled offend-
ers (n = 0) in this study ever reported a period of chronic homelessness. Conse-
quently, interchange—offenders alternating between periods ofchronic transiency
and stable, domiciled residency—did not occur between the two samples.
HYPOTHESES
Several classical school–positivist school debates remain when studying the
homeless(andcomparingthemtodomiciledpeople).Arethehomelessmoredan-
gerous than domiciled people? Are the homeless plagued by mental illness, or is
that a myth? Do homeless people abuse drugs and alcohol more than non-
homeless people? Are the homeless more dangerous, more violent, and more
prone to use weapons than domiciled people? Do homeless people have greater
criminality than domiciled people? This article compares homeless to domiciled
Criminality of Adult Homeless 61
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
jail inmates in a large adult county jail facility in Colorado. The current study’s
hypotheses are the following:
Hypothesis 1. Homeless people are morelikely to bearrested for violent offensesthan
domiciled people.
Hypothesis 2. Homeless people are more likely to be arrested for nuisance offenses
than domiciled people.
Hypothesis 3. Mental illness is more prevalent among the homeless than among the
domiciled.
Hypothesis4. Homelesspeoplearemorelikelytobealcoholicsthandomiciledpeople.
Hypothesis 5. Homeless people are more likely to be addicted to illicit drugs than
domiciled people.
Hypothesis6. Homeless people havemore extensivecriminal histories than domiciled
people.
Hypothesis 7. The criminal histories of homeless people will include more arrests for
crimes of violence, crimes involving the use of a weapon, and crimes involving
drugs and alcohol than domiciled people.
DATA AND METHODS
Official andself-reportcriminalhistory datawereusedfrom apretrialservices
unit at a large adult county jail in Colorado. Official data include national, state,
and local records measuring all of the respondents’ recorded criminal history.
Self-reportedcriminalhistoriesweretakenfromofficialbondinterviewsbetween
the pretrial services unit and arrestee-respondents. Interviews were conducted
from January through September of 1998. Due to severe underreporting of prior
arrest histories(see Hindelang, Hirschi, &Weis, 1979), self-reporthistories were
used only to supplement the more valid official arrest records.
SAMPLING
Sinceitsinceptionin1989,thepretrialservicesunitinthisstudyhasconducted
48,883 bond interviews. Due to multiplicity, the sampling frame included about
40,000 different arrestees.Two independentsimple-random samples yielded 100
homelessrespondentsand100domiciled respondents.Thissamplesize(N=200)
was selected because it was sufficiently large, under the Central Limit Theorem,
toaccuratelyportray theparameters of thetotal criminaloffender population (see
Babbie, 1995, pp. 195-203). About 5% to 15% of the 40,000 valid cases were
homeless; the remaining majority were domiciled arrestees. Records were avail-
able forall 200 respondents; thus,none were omittedfromthe study. Datacollec-
tion occurred in September 1998.
62 International Journal of Offender Therapy and Comparative Criminology
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
MEASUREMENT
Six demographic and six criminal history variables were used. Mental illness,
alcoholism, and drug addiction were determined by official records indicating
mental illness/criminal insanity and whether respondent was a registered drug
offender, as well as respondents’ self-reported treatment for chronic mental ill-
ness, alcoholism, and drug addiction and referrals from local mental health and
substance abuse treatment facilities. Crimes of violence included harassment,
assault, sexual assault,child abuse,brawling,publicfighting,and disorderly con-
duct. Nuisance crimes included all municipal ordinances often associated with
peopleof transientstatus.Examples ofmunicipaloffensesincludedpossession of
alcoholin public,camping withouta permit,aggressivebegging,vagrancy,use of
offensive words in public, indecent exposure, and public intoxication. Property
crimes included forgery, any theft or larceny, burglary, trespassing, and any type
of fraud. Traffic crimes included driving under the influence of alcohol, driving
without a license or insurance, vehicular eluding, habitual traffic offender, and
driving under restraint. Drug crimes included possession or sale of marijuana;
hashish; hallucinogens such as psilocybin mushrooms and LSD; powder and
crack cocaine; heroin; methamphetamine; and inhalants such as paint, glue, and
other noxious chemicals. Weapons offenses included the use of any firearm,
knife,bat,andmartialartsweaponry.Demographicandcriminalhistoryvariables
with coding appear below.
Sex: males (0), females (1)
Race: White (0), Hispanic (1), Black (2), Native American (3)
Age at current arrest: 18 through 79
Mentally ill: No (0), Yes (1)
Alcoholic: No (0), Yes (1)
Drug addict: No (0), Yes (1)
Current arrest type: Violence (0), Nuisance (1), Property (2), Traffic (3), Drug (4)
Number of prior arrests: 0 through 108
Number of prior arrests for violence: 0 through 27
Number of prior arrests for alcohol: 0 through 27
Number of prior arrests for drugs: 0 through 11
Number of prior arrests for weapons: 0 through 9
DATA ANALYSIS
Because ofdichotomous or dummy dependentvariables (likelihoodof violent
arrest, likelihood of nuisance arrest, presence of mental illness, presence of alco-
holism, and presence of drug addiction), logistic regression was used. Ordinary
least squares regression was used for the continuous criminal history dependent
variable.Becauseofpaucityofarrestsandresultantinsufficientsamplesize,com-
Criminality of Adult Homeless 63
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
parison ofmeans wasused totesthypotheses comparingprior arrest forcrimes of
violence, crimes employing weapons, and crimes involving drugs and alcohol.
RESULTS
Substantial differences existed between the homeless and domiciled arrestees
in this study. The average homeless jail inmate was a 33-year-old White male.
Most homeless jail inmates were not mentally ill or addicted to drugs; however,
almost half (43%) were alcoholics. The average homeless jail inmate had a sub-
stantial criminal history, including nearly 19 priorarrests. The average domiciled
jail inmatewas a30-year-old male.A majority of the domiciledjail inmates were
White (59%); however, a largepercentage (32%) were Hispanic. Most domiciled
arrestees were neither addicted to drugs or alcohol, nor were they generally
plagued by mental illness. More than half (53%) of domiciled jail inmates were
arrestedfor trafficchargessuch asdrivingunderthe influence.Foranoverviewof
the descriptive statistics of homeless and domiciled jail inmates, see Table 1.
Homeless jail inmates were 2% more likely than domiciled jail inmates to be
arrested for a violent offense (e
0.02
= 1.02). This difference was not significant.
Mentally ill jail inmates were significantly more likely (278%) than non–mentally
ill jail inmates to be arrested for a violent offense (e
1.33
= 3.78). Alcoholics were
significantly less likely (59%) than nonalcoholics to be arrested for a violent
offense (e
–0.89
= 0.41).
Homeless jail inmates were significantly more likely (169%) than domiciled
jail inmates to be arrested for a nuisanceoffense (e
0.99
= 2.69). Mental illness was
more prevalent among homeless jail inmates than among domiciled jail inmates.
Indeed, homeless arrestees were 222% more likely to suffer from mental illness
than domiciled arrestees (e
1.17
= 3.22). Of the homeless sample, 12% suffered
from mental illness. Age demonstrateda significant positiverelationshipto men-
tal illness.
Homeless jail inmates were not more likely than domiciled jail inmates to be
drugaddicts.However,nearlyhalfofthehomelesssample(43%)werealcoholics,
compared to 28% of the domiciled sample. The homeless also had significantly
more prior arrests for alcohol offenses than domiciled people. Relatively few
homeless(8%)ordomiciledpeople(7%)werearrestedfordrugcrimes.Similarly,
only 15% of the homeless sample and 12% of the domiciled people were regis-
tered or self-reported drug addicts. Homeless people averaged about one prior
drug arrest, whereas domiciled people averaged less than half a prior arrest for
drugs. Age demonstrated a significant relationship to being alcoholic. Each year
of aging increased the likelihood of being alcoholic by 7%.
Homeless arrestees were more likely than domiciled arrestees to have an
extensive prior arrest history. Predictably, age was positively related to criminal
history. Olderarrestees had moreextensivecriminal histories thanyounger arres-
tees. Being alcoholic was significantly and positively related to criminal history.
64 International Journal of Offender Therapy and Comparative Criminology
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
Theaveragecriminalhistory ofthe homelesshad4 timesas manyprior arrestsfor
violence than the average history of domiciled arrestees. Furthermore, homeless
people had significantly more prior arrests for use of weapons than domiciled
people. See Table 2 for difference in means between homeless and domiciled
offenders.
Thecriminal historiesof thehomeless offendersabsolutely dwarfedthecrimi-
nal histories of domiciled jail inmates. The average domiciled person had 4 prior
arrests; the average homeless person had nearly 19 prior arrests. The range of
priorarrestsfor domiciledpeople was0to 26.Therangeofpriorarrests forhome-
less people was 0 to 108.
Regression analyses indicated that the homeless jail inmates were more likely
thandomiciledjailinmates tobearrested fornuisancecrimes, weremorelikelyto
be mentally ill, and had more extensive criminal histories. Regression analysis
yielded no differences between homeless and domiciled jail inmates in terms of
Criminality of Adult Homeless 65
TABLE 1
DESCRIPTIVE STATISTICS: HOMELESS AND DOMICILED OFFENDERS
Variable Homeless Domiciled
Age (M) 32.9 30.5
Sex (%) 90 male 80 male
10 female 20 female
Race (%) 78 White 59 White
15 Hispanic 32 Hispanic
5 Black 7 Black
3 Native American 2 Native American
Alcoholic (%) 43 28
Drug addicted (%) 15 12
Mentally ill (%) 12 4
Prior arrests (M) 18.6 4.2
Prior arrests (range) 0-108 0-26
Alcohol arrests (M) 3.8 1.2
Alcohol arrests (range) 0-27 0-10
Drug arrests (M) 1.1 0.5
Drug arrests (range) 0-11 0-8
Weapon arrests (M) 0.5 0.2
Weapon arrests (range) 0-9 0-5
Current arrest (%):
Violence 22 21
Nuisance 27 11
Property 25 8
Traffic 18 53
Drug 8 7
NOTE: N = 200 (100 homeless and 100 domiciled).
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
66 International Journal of Offender Therapy and Comparative Criminology
TABLE 2
DIFFERENCE IN MEANS FOR HOMELESS AND DOMICILED OFFENDERS
T Value Levene’s Test
Age 1.48 1.35
Race 2.15* 5.02*
Sex 1.99* 16.84**
Alcoholic 2.23* 17.10**
Mentally ill 2.10* 19.13**
Drug addicted 0.62 1.54
Prior arrests 5.28** 86.19**
Alcohol arrests 3.77** 47.88**
Violent arrests 4.27** 44.20**
Weapon arrests 2.37** 16.21**
Arrest type 2.81** 1.71
NOTE: N = 200 (100 homeless and 100 domiciled).
*p < .05. **p < .01.
TABLE 3
LOGISTIC REGRESSION COEFFICIENTS FOR VIOLENT
OFFENSES (1), NUISANCE OFFENSES (2), MENTAL ILLNESS (3),
ALCOHOLISM (4), AND DRUG ADDICTION (5)
Independent Variable 1 2 3 4 5
Age 0.02 0.01 0.04* 0.07*** –0.01
(0.02) (0.02) (0.02) (0.02) (0.02)
Race –0.04 0.28 0.24 0.15 0.14
(0.27) (0.27) (0.36) (0.23) (0.31)
Sex –0.82 –1.96 –0.01 –0.85 –7.49
(0.60) (1.05) (0.81) (0.51) (18.15)
Homeless 0.02 0.99** 1.17* 0.48 0.10
(0.37) (0.41) (0.61) (0.33) (0.43)
Mentally ill 1.33* –0.60
(0.59) (0.83)
Alcoholic –0.89* 0.35
(0.44) (0.43)
Drug addicted 0.14 –0.80
(0.53) (0.68)
Chi-square 13.16 17.99 8.48 27.90 9.82
(pseudo) R
2
0.06 0.08 0.04 0.12 0.05
NOTE: N = 200 (100 homeless and 100 domiciled).
*p < .05. **p < .01. ***p < .001.
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
being arrested for a violent crime, being alcoholic, or being addicted to illicit
drugs. See Table 3 for regression results.
Comparisonofmeansanalysisindicatedthat,withtheexceptionofdrugaddic-
tion prevalence, all hypotheses were supported by the data: The homeless were
significantly more dangerous, violent, prone to use weapons, alcoholic, mentally
ill, and criminally active than domiciled jail inmates.
DISCUSSION AND CONCLUSION
In this study, homeless and domiciled jail inmates were equally likely to be
arrested for a crime ofviolence. This supported prior research (Zapf et al., 1996).
The criminal histories of the homelessincludedsignificantly more crimes of vio-
lence. The homeless in general were more likely to use weapons than domiciled
people. Perhaps, exposureto the streets makes the homeless employ weapons for
survival and self-defense.
Homeless offenders were more likely than domiciled offenders to be arrested
fora nuisancecrime. Indeed,the largestgroupof homelessoffendersin thisstudy
(27%) were arrested for nuisance crimes. However, homeless offenders were
nearly as likely to be arrested for property crimes such as burglary (25%) and
crimes of violence (22%). This contradicts Irwin (1985), who reported that jails
were filled with rabble incarcerated for harmless public order crimes. Mental ill-
ness was more prevalent among homeless jail inmates than among domiciled jail
inmates,althoughonly 12%of thehomelesssample sufferedfrom mental illness.
This seems to indicate that although mental illness is not characteristic of home-
less people in general (Simons et al., 1989; Snow et al., 1986), mental illness is
prevalent among alcoholic homeless with mammoth criminal histories. This
notion affirms Snow et al.’s (1986) idea that the heightened visibility (via a can-
tankerous public arrest) of mentally ill homeless men creates a public perception
that all homeless people are mentally deranged drunkards.
Drug addiction was not prevalent among homeless or domiciled offenders in
this study. This was counter to Zapf et al. (1996), who found that one third of
domiciled offendersandone fifthof homeless offenderswere addicts arrested for
drug crimes. Alcohol seemed to be the drug of choice for all members in this
study: More than half of domiciled people were arrested for traffic offenses such
as driving under the influence of alcohol.
These results should be considered with caution for three reasons. First, the
sample was limited to Colorado and not generalizable to other regions. Second,
the racial composition of this sample was disproportionately White (68%) and
Hispanic (24%). Perhaps, the criminal differences between homeless and domi-
ciledoffendersisdifferent forother racialandethnic groups.Similarly,the subur-
ban homeless and domiciled offenders in this study might be significantly differ-
entincriminal offendingand affective disordersthan urbanor rural homelessand
domiciled offenders. Third,the homeless represented about5%to 8% of thetotal
Criminality of Adult Homeless 67
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
jail inmatepopulation (and 5% to15% of the totalarrest population) in thisColo-
rado sample. Perhaps, the criminal activity of the homeless offenders is different
incities wherethehomeless makeup alargerproportionofjailinmates (e.g.,Dal-
las, Denver, Portland, Atlanta, Los Angeles) than in cities where the homeless
composeanexceedinglysmallproportionofjailinmates(e.g.,Cleveland,Detroit,
Birmingham, St. Louis, Philadelphia) (U.S. Department of Justice,1998). Police
enforcement and community tolerance for the homeless probably vary with
criminal-homeless population size.
In this study, the homeless offenders were generally more violent, had more
extensive criminal records, were more prone to use weapons, and were more
likelytobearrestedfor nuisanceoffensesand tobeplaguedby mentalillnessthan
domiciled offenders. Counter to popular belief, a majority of homeless offenders
werenotdangerous,mentallydisturbedalcoholics,althoughtheircriminalitywas
significantlygreater thandomiciledpeople. Asizablecomponent ofthehomeless
population inthis study (all of themabout 40-year-oldWhitemales) were violent
alcoholics with documented histories of mental illness and extraordinarily
lengthycriminal histories. Nocausal predictionsweremade thatcould support or
refuteclassicalorpositivistexplanationsforhomeless criminality. However,13%
of the homeless offenders in this sample were not harmless nuisance offenders
unfairlyharassedby police(Aulette&Aulette,1987;Irwin,1985),butinsteadyet
another disenfranchised group in thissociety with a rather nefarious involvement
in crime. Criminal justice practitioners could benefit from this epidemiological
information by channeling specific elements of the homeless population to the
mostappropriate destination:(a)substance abusecare facilities,(b) mentalhealth
facilities, (c) detoxification centers, (d) homeless shelters, and (e) jails.
REFERENCES
Aulette, J., & Aulette, A. (1987). Police harassment of the homeless. Humanity and Society, 11,
244-256.
Babbie, E. (1995). The practice of social research. Belmont, CA: Wadsworth.
Belcher, J. (1988). Are jails replacing the mental health system for the homeless and mentally ill?
Community Mental Health Journal, 24, 185-195.
Benda,B.(1987).Crime,drugabuse,mentalillness,andhomelessness.DeviantBehavior,8,361-375.
Gibbons,D.(1971).Observationsonthestudyof crime causation. AmericanJournalof Sociology,77,
262-278.
Hindelang, M., Hirschi,T., & Weis, J. (1979). Correlates of delinquency: Theillusion of discrepancy
between self-report and official measures. American Sociological Review, 44, 995-1014.
Irwin, J. (1985). The jail. Berkeley: University of California Press.
McCarthy,B.,&Hagan,J.(1991).Homelessness:Acriminogenicsituation?BritishJournalof Crimi-
nology, 31, 393-410.
McCarthy, B., & Hagan, J. (1992). Mean streets: The theoretical significance of situational delin-
quency among homeless youths. American Journal of Sociology, 98, 597-627.
Simons,R.,Whitbeck,L.,&Bales,A.(1989).Lifeonthestreets: Victimizationand psychologicaldis-
tress among the adult homeless. Journal of Interpersonal Violence, 4, 482-501.
68 International Journal of Offender Therapy and Comparative Criminology
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
Snow,D.,Baker,S.,&Anderson,L.(1989).Criminalityand homeless men: Anempiricalassessment.
Social Problems, 36, 532-549.
Snow, D., Baker, S., Anderson, L.,& Martin, M. (1986).The myth of pervasive mental illness among
the homeless. Social Problems, 33, 407-423.
U.S. Department of Justice. (1998). Arrestee drug abuse monitoring program annual report on adult
and juvenile arrestees, 1997. Washington, DC: Government Printing Office.
Wright, J. (1988a). The mentally ill homeless: What is myth and what is fact? Social Problems, 35,
182-191.
Wright, J. (1988b). The worthy and unworthy homeless. Society, 25, 64-69.
Zapf, P., Roesch, R., & Hart, S.(1996). An examinationof therelationship ofhomelessness tomental
disorder, criminal behavior,and health care ina pretrial jail population. Canadian Journalof Psy-
chiatry, 41, 435-440.
Matt DeLisi
Graduate student, Department of Sociology
University of Colorado
Boulder, CO 80309-0327
USA
Criminality of Adult Homeless 69
© 2000 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at IOWA STATE UNIV on March 26, 2008 http://ijo.sagepub.comDownloaded from
... Various studies (Aulette & Aulette, 1987;Irwin, 1985) have pointed out that persons experiencing homelessness are victims of excessive police harassment, which may help to explain the disproportionately high rates of police detention among this population. DeLisi (2000) noted that the crime rate in persons living homeless is likely the result of the unjust criminalization of life on the street. Additionally, migration processes can also include relevant vulnerability factors that lead to an individual becoming homelessness. ...
... The interviewees in Buenos Aires had experienced a large number of police detentions and arrests. In the opinion of various authors, persons experiencing homelessness are victims of excessive police harassment, which may help explain their high arrest rates (Aulette & Aulette, 1987;DeLisi, 2000;Irwin, 1985). DeLisi (2000) notes that homelessness seems to be inherently criminogenic; thus, the high crime rate observed among persons experiencing homelessness may stem from police harassment and the increased possibility of being arrested and not from the commission of real crimes (Aulette & Aulette, 1987;Irwin, 1985). ...
... In the opinion of various authors, persons experiencing homelessness are victims of excessive police harassment, which may help explain their high arrest rates (Aulette & Aulette, 1987;DeLisi, 2000;Irwin, 1985). DeLisi (2000) notes that homelessness seems to be inherently criminogenic; thus, the high crime rate observed among persons experiencing homelessness may stem from police harassment and the increased possibility of being arrested and not from the commission of real crimes (Aulette & Aulette, 1987;Irwin, 1985). In fact, Longhi (2022) highlights that homeless people in Buenos Aires experienced an increase in criminalization by security forces during the period of restrictions aimed at curbing the COVID-19 pandemic. ...
Stressful Life Events among Women Living Homeless in Argentina
Article
Full-text available
  • Aug 2022
This paper analyses the types, timing and perceived causality of SLEs experienced during childhood and adolescence and throughout the lives of women experiencing homelessness in Buenos Aires (n = 72). A structured interview was used to collect the information. Findings showed that interviewees had experienced a high number of SLEs during their childhood and adolescence. These SLEs were qualitatively severe and experienced from very early ages. Interviewees had also suffered from a large number of SLEs throughout their lives, mainly before becoming homeless. Interviewees attributed their homelessness largely to their experience with SLEs in their family environment, primarily during their childhood.
... Contained within tersely worded legal definitions related to, for example, strong-arm and armed robberies, are multifaceted behaviors simultaneously associated with both the application of coercive violence and the goal of obtaining another's property (Cook, 1987; see also Luckenbill, 1980). burglary (DeLisi, 2000;Fischer 1992b;. Homelessness is thus a complex, criminogenic and victimization-prone life circumstance. ...
... Physical violence is an endemic problem for those experiencing homelessness (North et al., 1993). Although there is a paucity of highly detailed or focused examinations, extant scholarship does support the generalization that homelessness is both a uniquely criminogenic and victimization-prone life circumstance (DeLisi, 2000;Fischer, 1992a;1992b;. ...
... 19 Far more common among homeless adults, however, are behaviors associated with public order and status offenses-misdemeanor or petty crimes (Fischer, 1988;Gonzalez et al., 2018; see also Fischer, 1992b;Fox et al., 2016;Newburn & Rock, 2006;Snow et al., 1989). Low-level crimes typically associated with homelessness include disturbances of the peace and disorderly behavior-related offenses, alcohol and drug use-related infractions, stealing from businesses, i.e. shoplifting, as well as other crimes associated with survival behaviors, such as trespassing or breaking into vacant structures or businesses in search of shelter (Anderson, 1923(Anderson, /2016DeLisi, 2000;Fischer, 1988;Gonzalez et al., 2018;Snow et al., 2018). ...
... Several studies (Aulette & Aulette, 1987;Irwin, 1985) have pointed out that people living homeless are victims of excessive police harassment, which would go to explain the disproportionately high levels of detention by the police (DeLisi, 2000). DeLisi (2000) comments that the criminality of persons living homeless could be the outcome of an unjust criminalization of life on the streets. ...
... Several studies (Aulette & Aulette, 1987;Irwin, 1985) have pointed out that people living homeless are victims of excessive police harassment, which would go to explain the disproportionately high levels of detention by the police (DeLisi, 2000). DeLisi (2000) comments that the criminality of persons living homeless could be the outcome of an unjust criminalization of life on the streets. DeLisi (2000) considers that lack of housing is inherently criminogenic. ...
... DeLisi (2000) comments that the criminality of persons living homeless could be the outcome of an unjust criminalization of life on the streets. DeLisi (2000) considers that lack of housing is inherently criminogenic. Aulette and Aulette (1987) and Irwin (1985) also comment that high levels of criminality amongst persons living homeless may be due more to unnecessary police harassment and an increased likelihood of arrest, and not because any actual crimes are committed. ...
Involvement in the Criminal Justice System and Incarceration among Women and Men Living Homeless in Spain
Article
  • Oct 2021
This article studies in a sample of women and men living homeless in Spain the issues that are most closely linked to involvement in the criminal justice system and incarceration. The study was carried out in Madrid (Spain) with a representative sample of men living homeless (n = 158) and a sample of women living homeless of a similar size (n = 138). A structured interview was used to gather the information. Results show that people living homeless in Madrid have had a greater degree of involvement with the criminal justice system and show higher levels of imprisonment, both before and after finding themselves in homeless situations. People living homeless who were imprisoned presented factors of particular vulnerability: Periods of childhood and adolescence in dysfunctional family settings, more frequently situations of homelessness, worse physical and mental health conditions, and higher levels of alcohol and drug abuse.
... Rough sleeping inherently exposes people to high risk of interaction with the justice system, due to having to find safe places to sleep and engaging in 'survival crimes' (Walsh, 2003;Barak & Bohm, 1989;DeLisi, 2000). ...
... The broad nature of the question means that these legal issues encompass family law court issues, criminal offences against property, civil claims, or violent offences. It's important to note that research on homeless people's interactions the justice system finds that the majority of offences committed by homeless people are minor and for petty crimes such as shoplifting or property damage that are often better described as survival behaviours rather criminal behaviours (Walsh, 2003;DeLisi, 2000). Figure 20 outlines the proportion of respondents in selected cohorts. ...
... The broad nature of the question means that these legal issues encompass family law court issues, criminal offences against property, civil claims, or violent offences. It's important to note that research on homeless people's interactions with the justice system finds that the majority of offences committed by homeless people are minor and for petty crime, such as shoplifting or property damage, often better described as survival behaviours rather criminal behaviours (Walsh, 2003;DeLisi, 2000). Figure 20 outlines the proportion of respondents in selected cohorts. ...
... Hindelang et al., 1978). Victimization is therefore a significant concern for those experiencing homelessness (Ellsworth, 2019a); so too is criminality (DeLisi, 2000;Fischer, 1988;Fox et al., 2016;Roy et al., 2014). Crime-related behaviors are prevalent to such a degree among the unsheltered, that a homeless existence has been characterized as a criminogenic life circumstance (McCarthy & Hagan, 1991). ...
... However, criminality occuring among those who experience homelessness is more typically associated with status offenses and low-level or misdemeanor crimes (Fischer, 1988;Gonzalez et al., 2018). Criminal activities common among homeless adults include public order offenses, alcohol and drug use-related infractions, violations of city ordinances, trespassing, and disturbances of the peace (DeLisi, 2000;Fischer, 1988;Gonzalez et al., 2018;Snow et al., 1989). Criminal acts, such as stealing from stores, are frequently a subsistence-related behavior, while in other instances, illicit behaviors are likely to reflect a disproportionate prevalence of psychiatric comorbidities (Fischer, 1992; see also Fischer et al., 1986;Fischer et al., 2008;Fox et al., 2016;White et al., 2006). ...
Housing and Criminality: The Effect of Housing Placement on Arrests among Chronically Homeless Adults
Article
  • Nov 2022
This article examines the relationship between criminality and housing among a population of formerly homeless adults by comparing arrest rates before and after placement in a supportive housing program. Jail entry records reflecting the period of chronic homelessness preceding housing placement were matched with post-intervention records for 87 adult men and women. In the studied population, a significant drop in arrest rates occurred after entry into the supportive housing environment. Findings support the assertion that supportive housing allows lifestyles to stabilize, which in turn leads to reductions in criminality. This study also reveals a transitionary period between the onset of the housing intervention and the observed decrease in arrests. Arrests do not instantaneously cease to accumulate. Instead, jail entries gradually decline over approximately twelve to eighteen months post housing placement. Explanations for the emergence of this transitionary period are discussed.
... The second group consists of mentally ill individuals arrested for survival behaviours (e.g., shoplifting, failing to pay) due to their experience of disadvantage and limited access to necessary care and services. Consistent with these two subtypes, there is some evidence that many individuals with mental illness come into contact with the justice system for minor offences (DeLisi 2000;Etter et al. 2008). The third group is comprised of mentally ill individuals who abuse or are addicted to alcohol and/or other substances. ...
Lifetime prevalence of mental illness and incarceration: An analysis by gender and Indigenous status
Article
  • Dec 2020
Individuals with mental illness are overrepresented in custodial settings. We examine the overlap between incarceration and diagnosed mental illness in a population‐based cohort born in Queensland in 1990. Data were extracted when the cohort was 23 or 24 years old. The population included 44,952 individuals (6.3 per cent Indigenous Australians, 45.8 per cent male), of which 1.5 per cent (n = 690) had at least one custodial sentence, and 6.1 per cent (n = 2,723) had at least one inpatient mental health diagnosis. Most individuals (91.5 per cent) with a mental health diagnosis did not have a custodial sentence. However, a substantial proportion of individuals (33.6 per cent) with a custodial sentence also had an inpatient mental health diagnosis. When examined by gender and Indigenous status, clear patterns emerged. Indigenous Australians were overrepresented in both the mental health and prison systems. Females with a custodial sentence were more likely than males to have a mental health diagnosis. Our analysis highlights the vulnerability of individuals with mental illness within the prison system.
... In the general population, homelessness was found to be an important pathway to incarceration (H.R. Lamb, Lamb, & Weinberger, 2001). Surveys of inmates found offenders with mental illness to be more likely than other inmates to be homeless either at the time of, or the year before arrest (DeLisi, 2000;Ditton, 1999;McCarthy & Hagan, 1991). Being homeless has also been associated with a greater number of arrests (Gelberg, Linn, & Leake, 1988;H. ...
Individual factors associated with recidivism among mental health court program clients
Article
There is growing evidence that diversion to a mental health court program (MHC) can reduce recidivism rates and improve the quality of life of clients. However, there is less known about MHC client characteristics and factors associated with recidivism. Yet, this information would be useful to increase the effectiveness of these programs. Cross-sectional quantitative data were collected on MHC clients in three consecutive years. Of the 155 program clients that were successfully interviewed, only 154 were included in the analysis due to one non-consent to collect further data from their case manager. The purpose of this secondary analysis was to examine "What individual factors are associated with recidivism among MHC program clients?" This analysis specifically explored the association of sex, age, low functional ability, homelessness, court site, and criminal history. From the multiple logistic regression results, the increased risk of recidivism was found to be significantly associated with younger clients and a prior criminal history. The results of this study suggest programs tailored to young adults and repeat offenders may be areas that MHCs could potentially focus on to increase their effectiveness.
... The potential breadth of these legal issues must be acknowledged; it could encompass family law court issues, criminal offences against property, civil claims, or violent offences. Extant research on homeless people's interactions with the justice system finds that the majority of offences committed by people experiencing homelessness are minor or petty crimes such as shoplifting or property damage, and further, that many of these offences can be categorised as 'survival behaviours' (Barak & Bohm, 1989;DeLisi, 2000;Walsh, 2003). Further breaking down experiences of legal issues by homelessness categories (Table 8.2, below), those who reported that they slept most frequently in institutional accommodation were the most likely to report current legal issues, with 44.7% of this cohort responding affirmatively to the question. ...
The State of Homelessness in Australia's Cities: A Health and Social Cost Too High
Technical Report
Full-text available
  • May 2018
Since 2010, Australian homelessness services, largely operating in the inner city areas of Australian cities, have undertaken interviews with over 8,000 people sleeping rough or otherwise homeless in concentrated data collection efforts called Registry Weeks. First implemented by US homelessness services as part of campaigns to end homelessness, Registry Weeks aim to develop a register of those who are homeless in areas in which homelessness services operate using a common interview schedule. The purpose of the register is for those who are homeless to be known by name and for their housing, health and social needs to be recognised to facilitate the organisation of local services to assist people into permanent housing with necessary supports. Over the seven years that the VI-SPDAT has been administered (2010-2017) interviews have been conducted with 8,370 people experiencing homelessness across Australian capital cities and regional centres. The State of Homelessness in Australia’s Cities: A Health and Social Cost Too High report is the first analysis of the consolidated Registry Week data across Australia.
... Not only do these factors increase the risk of housing instability and homelessness, but if homelessness occurs then ex-offenders face an increased likelihood of re-incarceration. Indeed, a pattern linking homelessness with high rates of re-incarceration has been reported in a number of studies in the US, Australia and elsewhere (Carlisle 1996;De Lisi 2000;Metraux and Culhane 2004;Baldry et al. 2006;Geller and Curtis 2011;Payne et al. 2015). Interestingly though, Cobb-Clark et al. (2016) find no significant relationship between previous incarceration and the duration of homeless episodes once controlling for time-invariant heterogeneity. ...
Homelessness and Incarceration: A Reciprocal Relationship?
Article
Full-text available
Objectives Examine whether exits from incarceration lead to homelessness and whether homelessness leads to incarceration. Methods This paper uses a unique longitudinal dataset which follows disadvantaged Australians over 2.5 years and provides very detailed information on their housing circumstances. Although studies consistently report a positive association between incarceration and homelessness, little is known about the causal relationship between them. We advance in that direction by exploiting the longitudinal dimension of our data in two ways: (i) employing individual fixed effects models to deal with time-invariant unobserved heterogeneity; (ii) lagging key independent variables to minimise reverse causality issues. Results Our results show that homelessness does not increase the risk of incarceration. In contrast, incarceration does increase the probability that an individual will become homeless, but not immediately. Exploiting details of the accommodation calendar 1-24 months after release, we find a modest immediate effect of incarceration on homelessness (a 3 percentage points increase), which increases 6 months after release (to around 12 percentage points) and persists for a further 11 months with respondents most often staying in precarious housing arrangements (boarding houses or with friends with no alternative) rather than becoming literally homelessness. Conclusions Our study shows the importance of having adequate coverage for post-release programs to break the link between incarceration and homelessness. Specifically, we find that the critical period for ex-inmates starts 6 months after release suggesting that this may be the time when support programs are currently lacking and would be most efficient.
Revealing Homelessness Phenomenon In Malaysia: A Case Study In George Town, Penang
Conference Paper
Full-text available
  • Sep 2019
Homelessness is not a new phenomenon in Malaysia, but in recent years it has become an increasingly prevalent social problem and requires a serious observation of all parties. The homelessness issues have also attracted the attention of the community and scholars because its implications on humanitarian issues, especially social exclusion and social justice. This study is based on three objectives, namely: (i) documenting demographic background of the homeless people in George Town; (ii) investigating the factors that caused them to become homeless people; and (iii) exploring the challenges of life experienced by the homeless people. Data collection was conducted through questionnaires with 30 homeless respondents, while in-depth interviews were conducted with 15 homeless informants in George Town. Data from the questionnaires were analyzed using SPSS, and data from in-depth interviews were analyzed using content analysis. The results of this study found that personal factors and environmental factors as the main contributor to the homeless population in George Town. This study has the potential to contribute to the knowledge on the issue of social exclusion which applies to the homeless people in Malaysia and its implications for the sustainability of urban community life.
The Mentally Ill Homeless: What is Myth and What is Fact?
Article
Full-text available
  • Apr 1988
A recent paper in this journal (Snow et al, 1986) has described and analyzed “the myth of pervasive mental illness among the homeless.” I review the procedures followed in that paper and conclude that they produce lower boundary, not best guess, estimates; the lower boundary on the percentage of (Austin, Texas) homeless who are mentally ill is 10 to 15 percent. Independent data from homeless clinical populations in 16 large cities suggest a best guess on the rate of mental illness among the homeless at about one-third. This is broadly consistent with previous research findings.
Mean Streets: The Theoretical Significance of Situational Delinquency Among Homeless Youths
Article
Full-text available
  • Nov 1992
Contemporary sociological theories of delinquency emphasize background and developmental factors while neglecting adverse situational conditions. This study uses data from youths on the street and in school to test an integration of strain and control theories that spans background and situational factors. After background and street exposure variables are controlled for and after school and street samples are combined, there is consistent evidence of the effects of adverse situational conditions: hunger causes theft of food, problems of hunger and shelter lead to serious theft, and problems of unemployment and shelter produce prostitution. These findings broaden and increase theoretical understandings of street life and crime.
Correlates of Delinquency: The Illusion of Discrepancy between Self-Report and Official Measures
Article
This paper reviews the research literature concerning the extent to which studies of delinquency that use official records produce results compatible with studies of delinquency that use self-reports of adolescents. Particular attention is given to sex, race, and social class as correlates of delinquency. The notion that official and self-report methods produce discrepant results with respect to sex, race, and class is largely illusory. In reaching conclusions of discrepancy several techniques have been used in the literature; the most general is the assumption that self-reports and official data tap the same domain of behavior. When the domain limitations of self-reports are recognized (and other illusory techniques are abandoned), the conclusion of general consistency between self-reports and official correlates for sex, race, and class emerges. This consistency and other evidence from victimization surveys, studies of the reliability and validity of self-reports, and studies of biases in criminal justice processing, suggest that both official data and self-reports provide valid indicators of the demographic characteristics of offenders, within the domain of behavior effectively tapped by each method.
Life on the Streets Victimization and Psychological Distress Among the Adult Homeless
Article
  • Dec 1989
Life-style/exposure theory and research findings concerning the consequences of criminal victimization were used to generate a set of hypotheses concerning the causes and consequences of criminal attack among the homeless. The sample for the study consisted of 79 homeless people residing in a midwestern state. The results largely supported the predictions. Employment problems, substance abuse, and, to a lesser degree, a history of psychiatric treatment increased involvement in a life-style based on desperate survival strategies (e.g., panhandling, rummaging through dumpsters, collecting cans, etc.). High utilization of these survival strategies was associated with high risk of criminal victimization. Being the victim of criminal attack, in turn, reduced feelings of self-efficacy and increased psychological distress.
The Myth of Pervasive Mental Illness Among the Homeless
Article
  • Jun 1986
This paper calls into question the double-edged thesis that the majority of the homeless are mentally ill and that the streets of urban America have consequently become the asylums of today. We present data from a triangulated field study of nearly 1,000 unattached homeless adults in Texas that contradict this stereotypic imagery. We also suggest that this root image is due to the medicalization of the problem of homelessness, a misplaced emphasis on the causal role of deinstitutionalization, the heightened visibility of homeless individuals who are mentally ill, and several conceptual and methodological shortcomings of previous attempts to assess the mental status of the homeless. We conclude by arguing that the most common face on the street is not that of the psychiatrically-impaired individual, but of one caught in a cycle of low-paying, dead-end jobs that fail to provide the means to get off and stay off the streets.
Criminality and Homeless Men: An Empirical Assessment
Article
  • Dec 1989
This paper examines the relationship between criminality and homelessness by tracking a random sample of homeless males through the police department records of a large Southwestern city over a 27-month period. When compared with data on criminality in the general population of males within the city over the same period, these data show that while the homeless have a higher overall arrest rate, the majority of offenses for which they are arrested are for public intoxication, followed by theft/shoplifting, violation of city ordinances, and burglary. The findings also suggest that criminality among homeless men varies with time on the streets and contact with the mental health system. Drawing on ethnographic data, these findings are explained in part in terms of criminalization, stigmatization, and adaptation processes. The findings challenge the depiction of homeless men as serious predatory criminals, and suggest a number of theoretical and policy implications.
Crime, drug abuse, mental illness, and homelessness
Article
  • Jan 1987
In this project, 345 homeless persons were interviewed during the period from May, 1985 to May, 1986 in Richmond, Virginia. Data reveal that current afflictions (e.g., hallucinations) are related to major life events in the past (i.e., psychiatric hospitalizations). Further, certain experiences prior to adulthood (e.g., drug use) predict these major life events (e.g., crime). Thus, the data support a general “drift down” hypothesis as a useful explanation of homeless for future theory development. Some policy implications are given.