Journal of Youth and Adolescence

, Volume 36, Issue 3, pp 291–299

Youth Homelessness and Social Stigma

Authors

    • Yale Program on Recovery and Community Health, Centre for Mountain Health Services, Mental Health Rehabilitation
Original Paper

DOI: 10.1007/s10964-006-9100-3

Cite this article as:
Kidd, S.A. J Youth Adolescence (2007) 36: 291. doi:10.1007/s10964-006-9100-3

Abstract

Building upon previous exploratory qualitative research (Kidd SA (2003) Child Adol Social Work J 20(4):235–261), this paper examines the mental health implications of social stigma as it is experienced by homeless youth. Surveys conducted with 208 youths on the streets and in agencies in New York City and Toronto revealed significant associations between perceived stigma due to homeless status and sexual orientation, pan handling and sex trade involvement, and amount of time homeless. Higher perceived stigma was also related to low self esteem, loneliness, feeling trapped, and suicidal ideation, with guilt/self-blame due to homeless status having the strongest impact on mental health variables.

Keywords

Youth homelessnessStreet youthSocial stigmaSuicidality

The family histories of most homeless youth are troubled, often consisting of disrupted and abusive home environments. High rates of drug and alcohol abuse are found among the parents of street youth, as is parental criminality (Hagan and McCarthy, 1997; Maclean et al., 1999). Additionally, a high percentage of the families are on social assistance (Ringwalt et al., 1998), and disrupted families are common with few homeless youth reporting having lived with both biological parents (Hagan and McCarthy, 1997). Also common are reports of marital discord (Dadds et al., 1993), domestic violence, and more household moves involving frequent changes of school (Buckner and Bassuk, 1997). The majority of the research into the backgrounds of street youth has focused on physical and sexual abuse, rates of which are consistently high (MacLean et al., 1999; Molnar et al., 1998; Ringwalt et al., 1998). Histories of emotional abuse (Ringwalt et al., 1998) and neglect (Daddis et al., 1993) are also frequently reported. These negative home experiences are associated with a host of other problems including poor performance in school, conflict with teachers, and conduct problems (Feitel et al., 1992; Hagan and McCarthy, 1997; Rotheram-Borus, 1993).

All of the phenomena described above are understood to be different or deviate from the ideals of the ‘social norm,’ and having such experiences has the effect of placing the individual outside of the cultural models of ‘normalcy.’ This is supported by a vast literature on the topic of child abuse, neglect and other dysfunctional backgrounds indicating that children who have suffered in these ways feel isolated and ostracized, and perceive others as dangerous and rejecting [for reviews see Wagner (1997) and Kendall-Tackett et al.(1993)]. Having such abusive and disrupted childhoods initiates a process of stigmatization in which children are identified and labelled as different, and as their opportunities, social and otherwise, narrow due to the beliefs and actions of others (Tomlin, 1991). For many homeless youth, having these types of early experiences likely leaves them more vulnerable to negative experiences associated with social stigma on the streets, given research showing that stigmatization has a greater impact upon the self-esteem of persons who have been abused in childhood (Coffey et al., 1996; Crocker and Major, 1989).

Street youth face many dangers and sources of stress in their lives on the street. To support themselves, they engage in numerous activities including trying to find work, seeking money from family/friends, panhandling, prostitution, survival sex (sex for food, shelter etc.), dealing drugs, and theft (Greene et al., 1999; Hagan and McCarthy, 1997; Kipke et al., 1997). The difficulty of surviving on the streets is highlighted by the large number of homeless youth who regularly lack shelter and go hungry (Antoniades and Tarasuk, 1998). Moreover, street life presents numerous dangers and stresses in the form of physical and sexual assaults and other types of victimization (Whitbeck et al., 2000). Drug abuse is a common way of coping with these stressors (Adlaf et al., 1996) and addiction is a major problem among this population (Greene and Ringwalt, 1996). There is a high incidence of mental disorders among homeless youth, with depression, PTSD, suicidal behaviors commonly found (Kidd, 2004; Whitbeck et al., 2000; Yoder, 1999). Mortality rates for these youth have been found to be 12–40 times that of the general population (Shaw and Dorling, 1998), with suicide having been found as the leading cause of death (Roy et al., 2004).

Despite the powerful and pervasive social stigma faced by homeless youth, it remains an overlooked topic in the research literature with few exceptions (Schissel, 1997; Kidd, 2003, 2004) with more commentary found in the adult homeless literature (Boydell et al., 2000; Lankenau, 1999; Phelan et al., 1997). Only two studies were found which examined social stigma as impacts the mental health of homeless youths. These studies, both utilizing an exploratory qualitative design, highlighted the range of socially oppressive and discriminatory actions faced on a daily basis by young homeless and street-involved persons (Kidd, 2003, 2004). It was found that perception of discrimination based upon negative stereotypes was related to feelings of worthlessness, loneliness and social alienation, and suicidality. Such linkages between social stigma, depression, and suicidality have been previously found among other groups of adolescents who face high levels of discrimination (e.g. gay, lesbian, and bisexual youth) (Radkowsky and Siegel, 1997) and among adult homeless among whom stigma has been related to social isolation and a devalued sense of self (Boydell et al., 2000).

Hypotheses

Given the exploratory findings described above, it was hypothesized that in the present study greater levels of perceived social stigma would be significantly related to lower self-esteem, loneliness, suicidal ideation, and the experience of feeling “trapped” with the trapped experience being closely linked with suicidality among homeless youth (Kidd, 2004, 2006). Three street demographic factors were hypothesized to have a significant impact upon levels of perceived social stigma. Prior work that has suggested that youths who are involved in the sex trade (Kidd and Kral, 2002) and homeless persons who engage in pan handling (Lankenau, 1999) are stigmatized to a greater extent. It was expected, therefore, that the degree of involvement in pan handling and sex trade activities would be positively associated with perceived stigma in the present study. The third street demographic variable, the total amount of time the youth had been homeless, was also expected to be positively related to perceived stigma. While such a relationship has not been examined previously, it would seem logical to suggest that the longer a youth had been homeless the greater his or her amount of exposure to stigmatizing circumstances and perception of stigma.

Gender, sexual orientation, and ethnicity were also expected to be related to the degree of stigma perceived by the participants in this study. There is a large body of literature suggesting that gay, lesbian, and bisexual adolescents both experience and perceive markedly higher levels of socially oppressive views and practices (see Radkowsky and Siegel, 1997, for review). Similar findings have been consistently highlighted for female adolescents (Leadbeater and Way, 1996) and ethnic minority adolescents (Comer, 1995). While perceived stigma among homeless youths having these demographic characteristics has not been evaluated, there is evidence that females and gay/lesbian homeless youth face greater adversity and victimization on the street (Cauce et al., 2000; Cochran et al., 2002; O’Grady and Gaetz, 2004). Given these findings it was expected that, for homeless adolescents, female gender, gay/lesbian sexual orientation, and non-white ethnicity would be positively related to perceived social stigma.

Lastly, of the aspects of stigma measured (understanding of public perception, actions of public against self, self-blame/guilt due to stigmatized status, struggle against larger society), feelings of shame and guilt were expected to have the greatest adverse impact upon mental health, a finding noted among other populations (e.g. HIV-positive persons) (Berger et al., 2001). Self-blame was also expected to be higher among sex trade-involved youth and gay, lesbian, and bisexual participants given previous findings of generally higher self-blame/guilt among those groups (Kidd and Kral, 2002; Kruks, 1991). Examination of self-blame as it was related to gender, ethnicity, pan handling, and total time on the streets was exploratory, however, given a lack of relevant previous findings to inform hypotheses.

Methods

Sample

To participate in the study youth had to be 24 years of age or younger and they must have had no fixed address or were living in a shelter at the time of the survey. The age criteria reflects the street context where youths tend to group together who are in this age range, perceive one another and be perceived by others as ‘street youth,’ and is the age cutoff used by many youth agencies.

Two hundred and eight youths participated in this study that took place in agencies and on the streets of New York City and Toronto. Street interviews were done in a range of locations where homeless youths congregated and/or pan handled (e.g., sidewalks of streets with heavy pedestrian traffic, public parks). Agency interviews included a youth agency in New York providing ranging services for disadvantaged youth and two agencies in Toronto which provide a similar range of drop-in services targeting homeless youth, one of which focused on providing services for youth involved in the sex trade. One hundred youths were interviewed in New York (39 agency, 61 street) and 108 in Toronto (31 at each agency, 46 street). On the streets potential participants were approached and, if they met criteria, the study was described to them. In agencies, participants were either referred to the study by agency staff or independently approached the researcher. Although interviewed in different cities, the youths’ narratives and survey responses did not suggest any notable variation on the basis of their geographical location. Youths from New York City and Toronto are not considered separately given the lack of any significant differences between these groups on any of the independent or dependent variables measured. Participants were reimbursed with 20 dollars in restaurant coupons and 97% of those approached agreed to participate. The data used in the present study were derived from the quantitative survey component of an interview including both qualitative and quantitative elements.

Of the 208 participants, 122 (59%) were male, 84 (40%) were female, and 2 (1%) were transgendered (male to female). With respect to ethnicity, 56% were White, 12% Black, 12% Hispanic, 5% Native, 14% of mixed ethnicity, and the remainder varied. The ages of participants ranged from 14 to 24, with a median age of 20. The average age of the youths’ first experience of leaving/being thrown out of home was 15, with a mean level of education of 10.6 years. A substantial proportion (57%) reported having been homeless for more than 2 years, with 33% reporting continuous homelessness and 40% having had conventional housing 25% of the time. Most youth resided in street and/or squat locations (47%), with 26% couch surfing (temporarily residing in the homes of others), and 14% lived in shelters. Most youths reported some combination of income sources with pan handling (45%), dealing drugs (23%), a job (23%), and sex trade involvement (15%) appearing with the most frequency.

A total of 46% of the participants reported making at least one suicide attempt in home or street environments, with 78% of attempters reporting that they had made more than one attempt. Differentiated by gender, 55% of females and 40% of males reported at least one attempt (t=2.07; p < .05). Methods of suicide attempts were overdosing (42%), cutting with a sharp object (32%), hanging (15%), jumping from a height (7%), with miscellaneous remainders. Although low sample sizes of non-white ethnicities prevent meaningful statistical comparisons of attempt rates, casual examination does not suggest extreme discrepancies: white (n=117, 49%), mixed ethnicity (n=28, 39%), black (n=25, 36%), Hispanic (n=24, 50%), Native (n=11, 46%).

Measures

Demographics

Along with endorsement of gender and ethnicity categories, participants indicated sexual orientation using a 7 point likert-type scale with 1=straight, 4=bisexual, and 7=gay. Pan handling was a dichotic variable which indicated whether or not pan handling was their primary subsistence activity. Sex trade involvement was defined as having “had sex with someone to get money, food, a place to stay, drugs, or something else you wanted,” and was measured using frequency categories (1=never; 2=1 – 3 times; 3=4 – 10 times; 4=11 or more times) as is consistent with previous work (e.g., Simons and Whitbeck, 1991).

Social stigma

No stigma questionnaire was located which addressed homelessness, youth or otherwise. As such, a 12-item survey was developed using 7 adapted items derived from an inventory designed for persons with HIV, and 5 items developed from previous qualitative work in which youth described their experiences of social stigma (Kidd, 2004). The stigma scale is brief and, as such, subject to criticism regarding narrow operations. Time constraints, however, regarding the administration of the survey which included questions in numerous domains necessitated brevity, as a lengthier questionnaire would have likely compromised the validity of responses among this particular group.

Items derived from the HIV scale were those which demonstrated high factor loadings with several aspects of stigma among persons with HIV, with item alteration essentially involving replacing “HIV” with “homeless.” As with the HIV scale, each item response was formatted as a 4-point Likert-type scale (strongly disagree, disagree, agree, and strongly agree). Adapted items included those assessing (i) aspects of the experience of being stigmatized: “I have been hurt by how people have reacted to me being homeless,” (ii) self-blame: “I feel that I am not as good as others because I am homeless,” “I feel guilty and ashamed because I am homeless,” (iii) perception of public attitudes: “People seem afraid of me because I am homeless,” “Some people act as though it is my fault that I am homeless,” “Homeless people are treated like outcasts,” “Knowing that you are homeless, people look for things wrong about you”. Items derived from previous qualitative work include: “I have been insulted by strangers because I am homeless,” “Most people think that homeless people are lazy and disgusting,” “Homeless can’t get jobs because they are homeless,” “I have to fight against the opinions and values of society,” and “Homeless people are harassed by the police because they are homeless.” Cronbach’s alpha for these 12 items was α=.87.
Table 1

Intercorrelations between stigma scale items

Abbreviated item

1

2

3

4

5

6

7

8

9

10

11

12

1. Hurt by how people react

.64**

.30**

.38**

.36**

.34**

.43**

.33**

.31**

.37**

.42**

.33**

2. Insulted by strangers

 

.59**

.56**

.20**

.14*

.47**

.47**

.44**

.41**

.52**

.48**

3. People seem afraid of me

  

.58**

.16*

‒.02

.48**

.44**

.41**

.43**

.44**

.31**

4. People act like it is my fault

   

.28**

.21**

.45**

.42**

.46**

.47**

.54**

.44**

5. I am not as good as others

    

.64**

.24**

.25**

.10

.16*

.13

.06

6. Feel guilty and ashamed

     

.23**

.22**

.14*

.03

.10

.13

7. People think homeless are lazy

      

.51**

.42**

.40**

.49**

.44**

8. Homeless people can’t get jobs

       

.41**

.28**

.37**

.29**

9. Police harassment

        

.50**

.56**

.40**

10. People look for things wrong

         

.71**

.58**

11. Treated like outcasts

          

.67**

12. Fight against opinions and values

           

*p < .05; **p < .01.

Intrapsychic variables

Self esteem was assessed using the total score from 5 Rosenberg (1989) items, with responses ranging from 0 to 5 (α=.83). Loneliness was measured using 4 items from the UCLA Loneliness Scale. These items were found to have the highest loadings on the general loneliness factor in a factor analytic study of this scale (Russell, 1996) and participants responded using a four point scale (α=.87). The trapped variable is comprised of factor scores derived from a factor analysis of three scales—feeling trapped, helplessness, and hopelessness, all measured on a four point scale. Hopelessness was measured using the participants’ total scores on 3 items from Beck’s Hopelessness Scale (Beck et al., 1974). These 3 items were found to have the highest correlations with the general hopelessness factor (Aish and Wasserman, 2001). Items assessing the participants’ experience of being and feeling trapped were derived was derived from participants’ descriptions in the qualitative work (Kidd, 2004). These items were “I feel trapped,” “I feel like I don’t have any real choices” and, “I feel like I don’t have anywhere else to turn” (α=0.86). Two items (α=0.79), assessing the helpless component of the trapped experience, had high loadings on the helplessness general factor, and were derived from the brief form of Lester’s Helplessness Scale (Lester, 2001). Total scores on these three scales were found to load onto a single factor (Factor loadings: Hopelessness=.85, Trapped=.89, Helplessness=.88; accounted for 76% of variance), with individual factor scores derived using the regression method. Current suicidal ideation was assessed through a total score from a 4 item scale commonly used in studies of non-homeless youth (Lewinsohn et al., 1996): “I though about killing myself,” “I thought about death/dying,” “I felt that everyone would be better off if I was dead,” and “I thought about plans to kill myself,” answered on a 4 point scale (1=never and 4=all of the time); α=.87. This scale was logarithmically corrected to adjust for skewness.

Results

Factor analysis of stigma scale

Inter-item correlations ranged from .02 to .71 (see Table 1), indicating that multicollinearity did not have a substantial effect upon the analysis. Alpha factor extraction was used as it is a procedure that maximizes internal consistency and is therefore appropriate for inventory development (Gorsuch, 1983). Alpha extraction utilizing varimax rotation, performed on the 12 inventory items, revealed a two factor solution (see Table 2 for item—factor correlations). The variance explained by these two factors was 51%. Factor 1 appeared to be a generalized stigma factor, with the highest item loading (.83) being the item “Homeless people are treated like outcasts,” and the remaining items, ranging in item loading from .49 to .83 indicative of understandings of stigmatizing public perceptions, actions by the public based upon stigma (insults, harassment, biases in hiring), being hurt by social stigma, and having to “fight” against social stigma. The first factor accounted for 40% of the total variance. The second factor, comprised primarily of the two self-blame items (feeling guilty and ashamed; not as good as others), had minimal weightings upon the first factor (.04, .10) and high correlations with the second factor (.78, .76). The only item that appeared to have relatively equal weightings between both factors was “I have been hurt by how people have reacted to me being homeless” (.49 on factor 1, .42 on factor 2). Given the degree of separation between these two factors as defined by the two self-blame items, it was decided to add these two items separately to form a self-blame subscale (α=.78) with the remainder of the items added to form a general stigma subscale (α=.89) for the purposes of the multiple regression analyses (i.e., as opposed to employing the sum of all items in defining the “general perceived stigma” construct).

Table 2

Abbreviated item

Factor 1

Factor 2

1. Hurt by how people react to me being homeless

.49

.42

2. Insulted by strangers

.72

.25

3. People seem afraid of me

.65

.09

4. People act like it is my fault

.68

.23

5. Feel I am not as good as others

.10

.76

6. Feel guilty and ashamed

.04

.78

7. Most people think homeless are lazy

.63

.28

8. Homeless people can’t get jobs

.52

.28

9. Police harassment

.64

.09

10. People look for things wrong

.72

−.08

11. Treated like outcasts

.83

.02

12. Fight against opinions and values

.66

.02

Demographics and social stigma

The various influences of gender, ethnicity, sexual orientation, total time on the streets, as well as pan handling and sex trade involvement upon perceived social stigma were examined, with the hypotheses that these variables would all have significant relationships with the experience of stigma. Contrary to the hypothesis, female gender was not found to have a significant impact on either the self-blame component of stigma (t=.120, p=.905) nor the general stigma component (t=.136, p=.892). For the purposes of analyzing the relationship between ethnicity and stigma, two groups were created, composed of White youth (n=117), and youth of other ethnicities [n=91; comprised of Hispanic (n=24); Black, (n=25); Asian, (n=3); Native, (n=11); mixed, (n=28)]. There was a significant difference between groups, though findings were not consistent with hypotheses that minority youth would report more stigma. It was found that white youths reported greater general perceived stigma (t=2.67, p<.01), with no significant difference in report of self-blame related to stigma (t=−.280, p=.78). Due to the degree of skewness of the 7 point sexual orientation scale (skew=1.22), 4 groups were created: straight (n=115), some degree of bisexual orientation (2 and 3 on the scale, n=40), bisexuality (n=31), and primarily gay/lesbian sexual orientation (5–7 on the scale, n=22). A one way ANOVA indicated that while sexual orientation was not significantly related to general stigma (F=1.28, p=.281), it was, as hypothesized, significantly related to self-blame (F=6.99, p < .001). Examination of pan handling activity and sex trade involvement as they related to stigma indicated that, consistent with hypotheses, youth who pan handle report higher levels of general stigma (t=5.99, p < .001) though lower self-blame (t=−2.04, p < .05) and degree of sex trade involvement was not associated with general stigma (r=.03, p=.643) but was related to higher levels of self-blame (r=.28, p < .001). Lastly, the hypothesis that total time on the streets would be related to perceived stigma was supported with respect to general stigma (r=.25, p < .001), but not self-blame (r=−.03, p=.71).
Table 3

Summary of regression analyses of stigma variables

 

Self-esteem

Loneliness

Trapped

Suicidal Ideation

Variable

B

SE B

β

B

SE B

β

B

SE B

β

B

SE B

β

General Stigma

.08

.03

.17*

.08

.03

.20**

.03

.01

.23**

.02

.01

.17*

Guilt/Self-Blame

.34

.11

.22**

.41

.10

.29**

.09

.02

.26**

.08

.02

.22**

Note. Adjusted R2 values were .09 for self esteem (reverse scored), .15 for loneliness, .15 for trapped, and .09 for suicidal ideation.

*p < .05; **p < .01.

Intrapsychic variables

Simultaneous multiple regression analyses were used to examine the relationship between negative self-image and general aspects of stigma and the dependent variables self-esteem, loneliness, feeling trapped, and suicidal ideation (see Table 3). For self-esteem, the model was found to be significant (F=10.75, p < .001, Adjusted R2=.09 – all R2 values reported are Adjusted values), with both general stigma (β=.17, p < .05) and self-blame (β=.22, p < .01) significant as predictors. The same analysis was repeated for the measure of loneliness, and it was found that the model including the general measure of stigma and self blame was significant (F=19.01, p < .001, R2=.15), with both general stigma (β=.20, p < .01) and self-blame (β=.29, p < .001) significant as predictors. Looking at stigma and how it related to feeling trapped, it was found that the regression model for general stigma and self-blame was significant (F=18.60, p < .001, R2=.15), with both general stigma (β=.23, p < .01) and self-blame (β=.26, p < .001) significant predictors. Finally, suicidal ideation was examined as a dependent variable. It was found that, with respect to general stigma and self blame, that the overall model was significant (F=11.22, p < .001, R2=.09) with both general stigma (β=.17, p < .05) and self-blame (β=.22, p < .01) significant.

Discussion

Previous work has suggested that homeless youth face intense social stigma (Kidd, 2003, 2004; Schissel, 1997), with qualitative analysis of youths’ narratives on their experiences as stigmatized persons suggesting an adverse impact upon their mental health (Kidd, 2004). The present study represented an initial effort to quantitatively examine the relationships between social stigma as perceived by homeless youth and various demographic and mental health domains. This analysis indicated that experience of social stigma varied depending on ethnicity, sexual orientation, subsistence activity, and total time on the streets. Reported experience of stigma was also found to have significant relationships with low self-esteem, suicidal ideation, loneliness, and feeling trapped.

Due to an inability to locate an existing brief inventory that measures various aspects of stigma as it is experienced by homeless persons, an inventory was created using items derived from an existing stigma scale designed for persons diagnosed with HIV (Berger et al., 2001) and items derived from the results of a qualitative analysis of homeless youth’s experiences with stigma (Kidd, 2003, 2004). The result was a 12-item inventory measuring several aspects of perceived stigma including understanding of negative public perception, actions of public against self (i.e., insults and harassment), self-blame/guilt due to stigmatized status, and struggling/fighting against the opinions and values of larger society. This measure proved to be internally consistent with factor analysis findings suggesting the presence of two distinct components: one comprised of self-blame/guilt related to stigmatized status and the other a general stigma factor comprised of all of the other items.

Several general demographic characteristics were hypothesized to be related to youths’ experiences with social stigma. Given previous work that has indicated that female adolescents experience greater amounts of discrimination (Leadbeater and Way, 1996) and evidence that homeless girls are more disadvantaged financially (O’Grady and Gaetz, 2004) and more frequently victimized on the streets (Cauce et al., 2000), it was expected that female participants would report greater levels of social stigma relative to males. This was not found to be the case for either general or self-blame aspects of social stigma. Similar hypotheses were made with respect to ethnicity and sexual orientation, based on evidence that ethnic minority adolescents (Comer, 1995) and gay, lesbian, and bisexual adolescents (Radkowsky and Siegel, 1997) face higher levels of stigma relative to the general population and gay, lesbian, and bisexual homeless youth are more frequently victimized on the streets (Cochran et al., 2002). The inverse of the hypothesized relationship between stigma and ethnicity was found, with white youths reporting greater general stigma though no significant difference was noted in degree of self-blame. Sexual orientation did, however, emerge as having a significant relationship with stigma. As predicted, a linear relationship was found between the degree of bisexuality and gay/lesbian sexual orientation and the amount of guilt and self-blame as it related to stigma. It would seem that gay and bisexual youth engage in more self-blame in reaction to stigma based upon homeless status, perhaps reflecting previous evidence of their having poorer psychological and physical health relative to straight homeless youth (Cochran et al., 2001). A similar relationship was not found with general stigma. The findings that gender, ethnicity, and sexual orientation (with respect to general stigma) did not emerge as predicted might be explained by the specificity of the inventory, which indicated stigma solely due to homeless status. As one participant noted while filling out the survey: “People aren’t afraid of me because I am homeless. People are afraid of me because I am Black.” It may be the case, with the exception of sexual orientation and self-blame, that stigma associated with these demographic variables is occurring in an additive fashion in which homelessness-specific stigma is not perceived as being substantially different among these groups. Females and minorities, however, may experience additional challenges due to further stigmatization specific to the ways in which they are identified by others. Future work may serve to better delineate the implications of multiple sources of stigma.

Street demographic hypotheses regarding the impact of pan handling, sex trade involvement, and total time homeless upon perceived stigma were confirmed. Pan handlers publicly display their status as homeless persons and regularly face humiliating interactions with strangers and authorities. It was found that having pan handling as a primary source of income was strongly related to perception of general social stigma. Conversely, pan handling had a significant negative relationship with guilt/self-blame, confirming Lankenau’s observation in his ethnography that pan handlers find constructive ways of managing their stigmatized identities (Lankenau, 1999). Sex trade involvement was found to be related to self-blame but, contrary to the hypothesis, not general stigma, possibly reflective of the additional stigma ascribed to prostitution (Brock, 1998). Sex trade work may, however, be similarly impacted by the question of multiple bases for discrimination unlike pan handling, which is more clearly and closely associated with homelessness. Lastly, the total amount of time spent homeless was significantly related to general perceived stigma, but not self-blame. This is likely similar to the relationship between stigma and pan handling in that youth who are homeless for longer are exposed to greater amounts of social stigma, with the greater amount of time potentially allowing for adjustment to discrimination such that guilt and self-blame in response to stigma are reduced.

Consistent with previous qualitative analyses (Kidd, 2004), perceived stigma was found to have a significant relationship with low self-esteem, loneliness, suicidal ideation, and feeling trapped. Feeling trapped, a construct comprised of elements of helplessness and hopelessness, has emerged in previous work as being central to suicidality among homeless youth (Kidd, 2004, 2006). Of the variables noted above, perceived stigma was most strongly associated with loneliness and feeling trapped. While the cross-sectional nature of this study does not allow for an examination of directionality of these relationships, these findings suggest that the well-documented tendency of society to blame homeless persons for their predicament (Phelan et al., 1997) may be serving to further compromise the mental health of youth already grappling with myriad risks and challenges. The potential influence of stigma in the lives of these youth may extend to mortality given the relationships between the above variables and suicide, and the recent finding that suicide is the leading cause of death for this population (Roy et al., 2004).

As hypothesized, guilt/self-blame due to stigmatized status emerged as the component of stigma with the strongest relationships with the mental health variables measured in this study. It is likely that guilt/self-blame is reflective of the degree to which these youths’ stigmatized status is internalized, with the implication that the degree to which homeless youths internalize society’s negative view of them is a central aspect of the process through which discrimination affects mental health. Such a finding has been noted among other populations (Lee et al., 2002).

The association between stigma and low self-esteem is not, contrary to common belief, typically found among most stigmatized groups (Crocker, 1999; Crocker and Major, 1989; Pinel, 1999). Recently, theorists have emphasized within-group variability (Crocker, 1999; Pinel, 1999), the meanings people give to situations in which stigmatization might be occurring (Crocker, 1999) and the protective coping strategies such individuals employ (Crocker and Major, 1989). Crocker and Major (1989) proposed a number of mediating factors that may help to explain the strong impact social stigma appears to be having on many street youth. These mediating influences, as adapted to the circumstances of homeless youth, include the consideration that since these youth have not had the stigma of homelessness since birth they have likely not had as many opportunities to adapt and develop coping strategies related to that stigma. They are also a group that has been exposed to many of mainstream society’s beliefs about drug addiction, poverty, prostitution etc. before they ever identify with these characteristics. Thus, negative stigmatizing evaluations are going to be more salient since they may to a certain extent have internalized those beliefs and more readily apply them to themselves. Responsibility for the stigmatizing condition is likely involved. Homeless youth are stigmatized for reasons that are largely thought to be the responsibility of the person (poverty, drug addiction etc.) (Schissel, 1997). Such groups are stigmatized to a greater extent than those not thought responsible for their conditions (e.g., developmental disability). With respect to coping with stigma, homeless youth may have difficulty putting in place the protective mechanism of devaluing the standards against which they are criticized. Beliefs regarding physical appearance, being drug free, financial success, and education are very central to Western culture and, as such, are difficult to ignore and devalue. Additionally, the abusive pasts of many street youth have likely left them more vulnerable to negative social and emotional consequences of stigmatization (Coffey et al., 1996). Lastly, the stigma and social oppression experienced by homeless youth appears to be occurring, for most, at a very high level. Homeless youth speak of a multi-levelled and institutionalized discrimination that is probably one of the more extreme forms to be found in North America, and is a constant harassment that cannot be escaped by ‘going home’ (Kidd, 2004).

The findings of this study could be substantially extended through a more thorough examination of the various aspects of stigma including the impacts and interactions of multiple stigmas, interactions between stigma and various social and street contextual variables, and how youths adapt to stigma over time. Examining these variables longitudinally would also prove informative in establishing the directionality of the interrelationships between variables.

Conclusions

These findings indicate that homeless youths’ experience of stigma plays a major role in their mental health status and suicide risk level. It will likely prove important for interventions to address social stigma as it is perceived and experienced by these young people, exploring how these perceptions are affecting their mental health in various domains, and helping them to find ways to insulate themselves from the discrimination that they face. This may involve working on ways to replace internalized messages of guilt and shame with a more empowering understanding of the various factors underlying stigma and systemic discrimination. Finally, counsellors should be aware of the heightened salience of discrimination and stigma for some groups, including gay, lesbian, bisexual, and transgendered persons, youth involved in the sex trade, and how perceptions of stigma change over time as youth are exposed to ongoing discrimination.

Copyright information

© Springer Science+Business Media, Inc. 2006