A Qualitative Study of the Formation and Composition of Social Networks Among Homeless Youth

Abstract

Although social networks are essential for explaining protective and risk factors among homeless youth, little is known about the formation and composition of these groups. In this study, we utilized 19 in-depth interviews with homeless youth to investigate their social network formation, role relationships, housing status, and network member functions. Our findings reveal that the formation of these networks occurred in different ways including meeting network members through others or in specific social situations. The majority of social network members were currently housed and provided various functions including instrumental and social support and protection. Responses from participants provide valuable insight into the formation of social networks and potentially explain their subsequent involvement in risky behaviors.

Social networks are an important part of normative adolescent development. These groups consist of a set of relationships that link social actors (Beaford, Gongaware, & Valdez, 2000) and generally include those who are in close proximity to one another (Cairns, Leung, & Cairns, 1995a). Among adolescents in general, social networks are often homogeneous as youth often select peers who are similar to themselves in terms of age, sex, race, personality, and behavior (Cotterell, 2007; Ennett & Bauman, 1994; Haynie & Osgood, 2005; McPherson, Smith-Lovin, & Cook, 2001). During adolescence, youth spend increasing amounts of time with members of their social network. As such, social networks play an important role in adolescent socialization as they may introduce youth to both prosocial and delinquent activities and experiences (Haynie & Osgood, 2005).

Conventional adolescents are not the only ones who rely on social network members as sources of socialization and support; other marginalized individuals such as homeless youth also rely on social networks. In general, homeless youth tend to be very diverse in terms of demographic characteristics and social network composition. For example, some studies indicate that racial and ethnic minorities are over-represented among homeless youth (Cauce et al., 1994; McCaskill, Toro, & Wolfe, 1998; Owen et al., 1998) and it is estimated that approximately 20% of homeless youth are lesbian, gay, bisexual or transgender (LGBT) compared to 10% in the general youth population (National Coalition for the Homeless, 2009). Because of this diversity, the social networks of homeless youth tend to be heterogeneous in nature and they also consist of individuals from home as well as from the street (Ennett, Bailey, & Federman, 1999; Johnson, Whitbeck, & Hoyt, 2005). Social networks that include other homeless youth are more likely to engender risk because of the high rate of substance use, delinquency, and risky sexual behaviors found among these individuals (Hagan & McCarthy, 1997; Tyler & Johnson, 2004; Tyler, Whitbeck, Chen, & Johnson, 2007; Whitbeck & Hoyt, 1999). Homeless youth who participate in these activities generally have friends who engage in similar practices (Kipke et al., 1998).

A few studies have examined the role (either risk or protective) that social network members play in the lives of homeless young people (Ennett et al., 1999; Rice, Milburn, Rotheram-Borus, Mallet, & Rosenthal, 2005); however, little is known about the initial formation and composition of these groups. In order to understand the formation and composition of the social networks of homeless youth, we utilize 19 in-depth interviews to explore the fundamental dynamics of their social networks. That is, we investigate homeless youths’ social network formation, the role relationships that exist in these social groups, the housing status of the members, and the functions provided by them. Because many homeless youth engage in behaviors that are detrimental to their health and well-being, we also asked homeless youth if they would like to change anything about their network members. Answers to these questions will provide valuable insight into the formation of social networks and potentially explain homeless youths’ subsequent involvement in high-risk behaviors. This information is important when designing prevention strategies for this group of at-risk youth.

Literature Review

The Formation of Social Networks

The term social network refers to the range of social relationships that are available to an individual. Social networks among homeless youth are generally comprised of people with whom an individual regularly associates and spends the majority of their time (Tyler, 2008). Although some studies do not specifically define social networks but instead focus on asking housed or homeless youth about their ‘friends’ (Haynie & Osgood, 2005; Rice et al., 2005), other works provide broad definitions which focus on people in the lives of homeless youth that they can count on for companionship, guidance, and support (cf. Johnson et al., 2005; Milburn et al., 2005; Smith, 2008). Despite the increasing body of literature on the social networks of homeless people (Johnson et al., 2005; Rice et al., 2005; Tyler, 2008), little is known about the formation of these groups. For example, Smith (2008) conducted in-depth interviews and explored the formation of “street families,” which refers to the self-supportive networks of homeless youth. She found that many youth suggested that the street families naturally emerged from a shared sense of the homelessness experience (Smith, 2008). As such, their commonality is their shared social circumstance of being homeless (Tyler, Melander, & Almazon, 2010). These relationships may emerge due to physical propinquity as homeless youth are likely to form ties with those who are in close proximity to themselves (Cairns et al., 1995a). Physical propinquity may also explain why the social networks of homeless youth tend to be heterogeneous (Johnson et al., 2005; Kipke et al., 1997; Rice, Milburn, & Rotheram-Bours, 2007), encompassing a wider range of individuals in terms of age, sex, role relationships, and/or housing status (Ennett et al., 1999; Johnson et al., 2005) compared to those of general adolescent samples (Cairns et al., 1995a; Cotterell, 2007). Finally, another key feature of homeless youths’ social networks is that they tend to be smaller on average than those of other adolescents (Cairns, Leung, Buchanan, & Cairns, 1995b). For example, Ennett and colleagues (1999) found that the average size of social networks of homeless youth was small (i.e., 2.6) whereas Cairns et al. (1995b) found the average size of seventh graders to be approximately 4.1 and Haynie and Osgood (2005) found that youth, on average, reported that they had 5.7 friends in their social network, using the National Longitudinal Study of Adolescent Health.

Relationship with Social Network Members and Their Housing Status

The relationships that homeless youths have with the members of their social network may also vary. Homeless youth often report that they associate with friends. Johnson et al. (2005) found that 71% of the homeless youth in their sample had a friend in their social network. Others have found that homeless youths’ social networks also include family members. For example, Tyler (2008) found that 12% of homeless youth reported having a family member in their social network. Romantic partners are also often named as people in homeless youths’ social networks: 52% of homeless youth reported that a significant other was a member of their social network (Usborne, Lydon, & Taylor, 2009).

Housing status may also impact the functions and activities of social networks. Some researchers have found that these groups are generally comprised of both homeless and housed individuals (Milburn et al., 2005; Montgomery et al., 2002). For example, Johnson and colleagues (2005) who distinguished between social networks from “home” versus those from the street found that although 47% of their sample reported that their social networks were comprised of one or more homeless individuals, 78% had at least one housed person in their network. Overall, the role relationships of members within their social networks appear to be heterogeneous and these groups are comprised of individuals from both home and the street.

Functions of Social Network Members

Although homeless youths may engage in risky and illegal activities with the members of their social network (Hagan & McCarthy, 1997), pro-social peers are often a hallmark of these groups (Rice et al., 2007). Social network members are sources of both emotional and material support (Johnson et al., 2005; Molina, 2000; Smith, 2008) and homeless youth generally report feeling close to their network members (Tyler, 2008). These social groups often supply companionship and moral support, which may mitigate homeless youths’ feelings of alienation and loneliness that they may routinely experience (Molina, 2000; Smith, 2008). Furthermore, network members are often instrumental in homeless individuals’ survival strategies as they may provide money and/or information on where to obtain food, clothing, or shelter as well as protection from victimization on the street (Auerswald & Eyre, 2002; Molina, 2000; Smith, 2008).

Social networks may also buffer against participation in risky behaviors. For example, having a family member in a homeless youths’ social network protects against risky sexual and/or and drug use behaviors (Ennett et al., 1999; Tyler, 2008). Conversely, some network members may actually encourage deviant behaviors. For example, Toro, Tulloch, and Ouelleette (2008) found that homeless adults who are in more supportive networks are more likely to abuse controlled substances. Similarly, Molina (2000) reported that homeless men often acquire and circulate illegal drugs and alcohol among their social networks. Although previous studies have not focused on things that homeless youth would like to change about their network members, it is possible that these areas of risk may be something these young people would like to modify. Consequently, we explored this topic in the current study.

Methods

The qualitative data for the present study are from the Social Network and Homeless Youth Project, a larger study designed to examine the effect of social network characteristics on homeless youths’ HIV risk behaviors. A total of 249 homeless youth (137 females; 112 males) were interviewed in shelters and on the streets from January 2008 to March 2009 in three Midwestern cities in the United States. Three experienced female interviewers conducted these quantitative interviews. These individuals were chosen because they have worked on past homeless youth projects, have served for several years in agencies and shelters that support at-risk youth, and are very familiar with local street cultures and know where to locate youth. Additionally, because two interviewers had previously worked at two of the sampled shelters and one interviewer was currently employed with one of the agencies, they were known and trusted by many of the participants. Furthermore, the interviewers routinely attended “group sessions” in the evenings with homeless youth, which further enhanced their rapport with the young people. All interviewers completed the Collaborative Institutional Review Board (IRB) Training Initiative course for the protection of human subjects in research.

Selection criteria for this study required participants to meet the definition of runaway or homeless and be between the ages of 14 and 21. The term runaway refers to youth under age 18 who have spent the previous night away from home without the permission of parents or guardians. Homeless youth are those who have spent the previous night with a stranger, in a shelter or public place, on the street, in a hotel room, staying with friends (e.g., couch surfing), or other places not intended as their resident domicile.

Participants for the qualitative interviews were selected from the original sample of 249 to represent different gender, racial/ethnic, and sexual orientation groups using a purposive sampling strategy (Miles & Huberman, 1994). After the completion of the survey instrument, interviewers selected youth from these different demographic groups to participate in an in-depth interview that was conducted approximately one week later. Interviewers were instructed to oversample racial/ethnic and sexual minorities because they are at greater risk for acquiring HIV (Centers for Disease Control and Prevention, 2002a; 2002b), which was the focus of the larger research project. All selected youth participated in these qualitative interviews. Interviewers gave the youth a card with their name and phone number along with the day and time for the in-depth interview. Youth were allowed to use shelter agency telephones to contact interviewers if they needed to reschedule the appointment. They were paid $30 for completing the qualitative interview which lasted approximately 1 to 1 ½ hours. All in-depth interviews took place in a private room at the shelters. Informed consent was obtained from all youth prior to the interview. Interviewers offered agency services or referrals to all youth (e.g., shelter, food services, and counseling). Participants were asked a series of open-ended questions, and all interviews were audiotaped and transcribed verbatim. Pseudonyms are used to preserve confidentiality. The university IRB approved this study.

Interviewer Guide

The guide for the semi-structured interviews consisted of a series of open-ended questions and probes that expounded upon topics in the quantitative survey. For the quantitative survey, youth could list up to five people that they see or spend most of their time with as well as three people they had sexual relations with in the past six months for a total of eight social network members. The sexual partners could be people on their original network list of five or new ones not previously mentioned. In either scenario, sexual partners listed are considered part of the youth’s social network. This approach has been used in past research on social networks and high-risk populations of similar age (Montgomery et al., 2002). The qualitative interviews began with the following statement: “Today I would like to talk with you in-depth about the same people that you told me about last time we did your other interview.” As a reminder, youth were then given a card with the initials of the people that they discussed in the survey. Specifically, interviewers asked respondents the following questions: “How did you meet each member of your network?” “Who made the first contact?” “What is your relationship to each network member?” “Where does each member live?” “What does each of your network members do for you?” and “What would you like to change about each network member?”

Although some network studies ask about individuals who provide specific resources such as instrumental and emotional support (Johnson et al., 2005), we were interested in focusing on the people with whom they spend the majority of their time. As such, we allowed the youth to define the boundaries of their social network by not having them restrict group membership according to age or relational criteria. In other young adult social network studies, the participants are only prompted to discuss their relationships with same-aged peers or those who they would consider their friends (Adamczyk & Felson, 2006; Haynie & Osgood, 2005). This does not reflect the lived reality of homeless youth as they are often forced to rely on unconventional others for support and companionship.

Participants

As indicated in Table 1, the qualitative sample included 13 females (68.4%) and 6 males (31.6%). Females tend to be slightly overrepresented among homeless youth (Whitbeck et al., 2004). Ten (52.6%) self-identified as gay (n = 2), lesbian (n = 1), bisexual (n = 6), and transgendered (n = 1). Because the interviewers were instructed to oversample sexual minorities, the numbers in this subsample are higher than what we would typically see in the population of homeless youth in general. For example, in the full sample, almost 18% self-identified as GLBT, which is consistent with previous studies (Whitbeck et al., 2004). Ages ranged from 16 to 21 with a mean of 19.47 years. The majority of the sample was White (n = 11; 57.9%) with the remaining respondents self-identifying as Black (n = 4), Hispanic (n = 2), American Indian (n = 1), and biracial (n = 1). On average, homeless youth in the full sample reported having 4.90 network members compared to 5.53 members in the qualitative sub sample, which is similar to that reported among housed youth. For example, Haynie and Osgood (2005) found that youth, on average, reported 5.7 friends in their network, even though they were allowed to nominate up to 10 individuals. Other key network characteristics are also presented in Table 1.

Table 1.

Descriptive Information for Full Sample and Qualitative Subsample

Variables	Full Sample (N = 249)		Qualitative Sample (N = 19)
Respondent Characteristics	N	%	N	%
Female	137	55.0	13	68.4
White	123	49.4	11	57.9
Black	59	23.7	4	21.1
Hispanic	20	8.0	2	10.5
American Indian	12	4.8	1	5.3
Asian	3	1.2	0	0
Biracial	22	8.8	1	5.3
Multiracial	10	4.0	0	0
GLBT	44	17.7	10	52.6
	Mean	SD	Mean	SD
Mean age	18.53	1.82	19.47	1.35
Network Characteristics	Mean	SD	Mean	SD
Mean total network size	4.90	1.56	5.53	1.65
Mean age of network members in years	24.30	6.55	26.01	5.01
aMean network stability	1.49	.54	1.64	.68
bMean frequency of network interaction	2.12	.67	2.24	.70
cMean network closeness	1.73	.52	1.75	.50
dMean frequency of network conflict	1.66	.41	1.66	.41
eNumber of network support from each member	9.56	4.46	9.16	5.70

^aResponses ranged from 1 (known for a year or more) to 5 (known for a few hours).

^bResponses ranged from 1 (saw everyday) to 4 (saw once or twice in the past month).

^cResponses ranged from 1 (very close) to 4 (not close at all).

^dResponses ranged from 1 (never) to 4 (always).

^eResponses ranged from 0 (no support from any member) to 25 (all types of support from all members).

Data Analysis

The interview transcriptions were imported into ATLAS.ti, a data management software program (Muhr, 2004). The first step in the preliminary data analysis involved rereading each interview transcript in its entirety in order to gain a deeper sense of the data as a whole. Because we were interested in the social network composition and formation among homeless youth, we then focused on the transcription sections that were related to the interview questions on this topic.

We assessed validity by triangulating the data by building evidence for a code or theme (e.g., how network members met) from several individuals (Creswell & Plano Clark, 2011). For intercoder agreement, we used a predetermined coding scheme and a qualitative codebook to identify whether we assigned the same or different codes between text passages (Miles & Huberman, 1994). In cases in which the intercoder agreement between the two authors was low or discrepancies existed, we obtained consensus through deliberation and re-evaluating our coding and themes. Table 2 presents sample quotes for each qualitative theme.

Table 2.

Social Network Formation Sample Quotes

Qualitative Codes and Subcodes	Selected Qualitative Quotes
Social network formation and roles
Met through others	- First met at my friend’s house. We were having a bonfire… all of our friends came and she happened to be there…
Met through normative social circumstances	- How did we meet? Well, it was a[t] school, and we were real close.
Met through shared circumstances	- R.D. I met through, I just met on the streets, um, got to talking, hanging out, and started dating.
First contact initiation
Respondent contacted network member	- Yeah I said hi. And then it went from there.
Mutual contact	- You know it was a sort of a mutual thing since we were introduced.
Network member contacted respondent	- She contacted me…First when I try to get in contact with her or anything she refused. But then she found out she knew my sister…
Unknown who made first contact	- Some people just came up and were just…“what’s your name”, “where you from?”
Role relationships with social network member
Current or former partner	- [We] obviously click…We’ve been in a relationship since May…
Family	- He treats me like a younger sister. He’s very protective, of everybody.
Friends	- He’s been my best friend since I was like fourteen, and currently he’s in jail but I visit him…
Other	- Well for me, I guess my relationship with her is kind of like, in a way she’s my mentor.
Social network members’ housing status
Housed	- They all live together, um, in a nice apartment that they can’t afford.
Homeless	- She [is] kind of homeless but like she’d never like stay out in the streets. She always finds somewhere to stay.
Other locations	- She’s in foster care right now, my little sister. She made some bad choices and she’s in foster care.
What network member does for respondent
Instrumental support	- He gives me places to stay and he will feed me and I’ve loaned him money before and it just kind of goes back and forth.
Social support	- We basically stand strong together. If you always have somebody by your side and somebody who knows what you’re going through, it’s easier to get through it.
Protection	- Everybody knows him so nobody would mess with me or fuck with me because of him.
Member does not help out	- Um S.W. doesn’t do a lot of anything for me, I don’t do a lot of anything for him.
What they would change about their network members
Personality characteristics	- She got one of those attitudes to where she’d pop off at the wrong person and the persons gonna say forget it and just lay her flat out, throw her in the hospital.
Lifestyle	- I mean I would really prefer her not to do meth, because I’m trying to stop…
Change their friends/partners	- I can’t stand her husband, he is so annoying and full of crap…
No changes to network members	- I don’t want to change my friends. I like them for who they are.

Findings

Social Network Formation and Roles

The formation and role relationships within homeless youths’ social networks were varied. The formation of these networks occurred in different ways including meeting network members through others or in specific social situations. Sometimes the homeless youths initiated contact with their network members whereas others reported that their network member made the first introduction or their initial encounter was mutual. Furthermore, the roles these network members occupied ranged from intimate partnerships to more peripheral relationships. Each of these factors impacts the form and function of social networks and is important to our understanding of the dynamics of these groups.

Met through others. In terms of meeting through other individuals, Elizabeth, a White female, discusses how she met one of her network members through a sister: I met through my foster sister…when I was in the foster [care] system. She was my foster sister in one of my foster homes and I was downtown one day, just holding a cigarette for my friend, and basically he started, he was like ‘I hope you’re not smoking,’ and I was like, ‘No, I’m just holding it for a friend,’ and, uh, he was like, ‘You look like you’re pregnant.’ I said, ‘I am, I’m about four months along,’ and he goes, ‘Well, that’s how far along my sister is,’ and I was like, ‘My sister’s four months pregnant, too,’ and then here comes my foster sister, walking down the street, and he goes, ‘Well, there’s my sister,’ and I’m like, ‘No, way’ that’s my sister,’ and he was like, ‘Well, we have something in common.’ [laughs]

Tyrell, a 19-year-old Black male also reported meeting his network member because she was one of his sisters’ friends. Another youth, Emily, a White heterosexual, was attending a bonfire with a group of individuals and met her network member via their friendship circle. Finally, Sarah, a White transgendered youth was introduced to her network member through a former partner. As such, individuals currently known to homeless youth were instrumental in introducing them to new network members.

Met through normative social circumstances

Respondents also reported that they met their social network members in more normative social situations. For example, some respondents such as Ashley met their network members at work. Others met these individuals at school: “I think we contacted each other in class. But we, we just actually started hanging out at the laundry mat, [be]cause her mom worked there” (Michael; 21-year-old gay male). Finally, some homeless youth reported that they have been acquainted with their network members for a long time. David and Tyrell, for example, have known their network members since childhood. As such, these respondents highlight that they contacted their network members in a manner that is similar to general adolescent populations.

Met through shared circumstances

For those who met because of their current life situation, it is likely that these youth formed an immediate bond because they were facing similar circumstances. For example, several youth reported meeting at least some of their social network members on the street. One young woman, Elizabeth, recalls, “R.D…I just met on the streets. Um, [we] got to talking, hanging out, and started dating.” Some youth discussed their first meeting with their network member at a shelter. Stephanie, a White, bisexual female said, “I think we met, we met at the shelter…first night that I was homeless, um, she was in the room next to me, and then we kept talking and then we kept running into each other.” Others met their network members while incarcerated (Darnel, black heterosexual) or in drug treatment (Megan, White lesbian). In these circumstances, the common situation of being homeless or being in a particular facility is what brought these network members together.

First contact initiation

Numerous youth reported that they initiated the first contact with the majority of their social network members. For many of them, it was as simple as saying “hi” to the other person and the conversation proceeded from there. For example, Michael recalls:

I did make the first contact. He [network member] was at the gas tank pumping gas and I was like ‘Damn’ [laughs] and that’s, um, and it kind of just went from there. I was like ‘Damn, what’s up with you?’ No, I just talked to…I don’t know, when I think someone’s cute, I tell them.

Others indicated that their first contact was mutual contact because they were introduced by a third party. This was the case for Stephanie who recalled that, “You know it was a sort of a mutual thing since we were introduced. So, I think…she’s the one who said the first words to me, which we, you know, ‘Hi, nice to meet you’.” In other situations, the network member initially made the first contact, even if they had a negative first interaction. Tyrell reports, “She contact[ed] me…First when I tr[ied] to get in contact with her or anything she refused. But then she found out she knew my sister so…” In other words, the relationship between Tyrell and his network member formed due to their mutual acquaintance. The involvement of the third party (Tyrell’s sister) made this interaction possible and eventually led to network ties between Tyrell and his friend. Without this mutual association, the currently housed network member may have been reluctant to interact with the homeless youth. Another youth reported that her current romantic partner made the initial contact. Elizabeth recalls, “And we just, you know, basically hit it off and started dating, and then I moved in with him.” For a young woman living on the streets, the offer of comfort and security that comes with living in an apartment as opposed to staying at a shelter may make the decision to move in with a partner more attractive and simple.

For other individuals, it was unclear who initiated contact. Instead, youth reports suggest that they asked their network members a series of questions in order to learn more about them before they decided whether or not to establish a relationship. For example, Melissa, a White heterosexual, recalls that, “some people just came up and were just…‘what’s your name?,’ ‘where you from?,’ [and] ‘did you just move here?’ Questions like that.” This quote suggests that these youth may want to get to know one another to some extent before deciding if this was someone with whom they would want to connect.

Role relationships with social network members

The network members were diverse in terms of age and role relationships which reinforced the heterogeneity of homeless youths’ social networks. Some network members were under the age of 5 (e.g., younger siblings, a child) whereas other members were in their late fifties (e.g., grandparents). Respondents also used a variety of terminology to refer to their relationship with each of their network members. These were grouped into four main subthemes: current or former partner, family, friends, and other.

Current or former partner

Respondents routinely listed committed, casual, or former partners in their current social networks. For example, some homeless youth reported their network members’ status as that of boyfriend, girlfriend, or fiancé. For example, Megan describes her current relationship:

Well I obviously click with her [fiancé] very well. We’ve been in a relationship since May, and um she proposed to me about a month ago, on our six month anniversary. I just really like um, the way she goes about things. She’s really spunky and she likes to have fun.

Other homeless youth spoke about network members being casual partners with whom they were dating on and off. For example, Amanda, a White bisexual female, reflected on her sex partner’s role relationship: “[sighs] Wow, that’s a tough one. We’re together, but not together, if that makes any sense to you? Like [we are] kind of dating, but kind of not [dating].” Others reported continuing contact with former partners. For example, Jennifer indicates that one of her members is “an ex-boyfriend but we’re still close.” Brittany, a bi-racial, bisexual female referred to one of her network members as an “associate” which to her meant, “We wasn’t boyfriend and girlfriend; we were just, like, associates but with pleasure, in a way.” Although some youth were in relationships that were more serious and of a longer duration, others indicated that their network members were casual partners that lacked commitment.

Family

Although some may assume that the social networks of homeless youth are comprised entirely of street youth, this is typically not the case. Youth generally listed family members other than parents or caretakers as being part of their social network. Michael spoke about his brother, a current member of his social network: “He actually plays a really, really, really big role in my life [be]cause he’s my little brother.” Other homeless youth spoke about network members as fictive kin (i.e., they assigned family status even though they were not related). Stephanie, a White, bisexual female, reflected on her network member: “He’s [network member] a lot like a guardian to me ‘cuz…’ he’s almost like a big brother-guardian sort of person…” Additionally, Jennifer, a White bisexual female, said the following about her network member: “He treats me like a younger sister. He’s very protective, of everybody.” Although the two examples above reveal that Stephanie and Jennifer were not related to their network members, they assigned family status to these two males because of the protective role they played in these women’s lives.

Friends

The third category of role relationships that we found included that of friends, and a majority of social network members fell within this grouping. Maria, a Hispanic bisexual, mentioned, “Right now, you know, he’s a friend. When I need to talk to him, I just chat him up on Yahoo and we just talk for hours.” Megan says this of her best friend, who is a member of her social network: “He’s been my best friend since I was like 14, and currently he’s in jail but I visit him, and send him money, and send him phone cards so he can call me and stuff, so we’re really close.” Tyrell referred to his network member as a “friend with benefits,” suggesting overlap between friendship and sexual intimacy. According to Tyrell, “Well it’s kinda what every man wants; friends with benefits.” This quote illustrates that the friendships of some homeless youth may not be completely platonic.

Other

The final category of role relationships included social network members who served as mentors or individuals whose relationship to the homeless youth was unknown. For the latter category, this person was someone in the network that they became intoxicated or “high” with but with whom they did not have a formal relationship. For example, Darnel reported, “The only time we’re hanging around each other is when we’re smoking weed or getting high or smoking meth or drinking or whatever.” In other cases, young people did not assign a role relationship to a person that they previously mentioned in the quantitative interview, which was conducted approximately one week prior to the in-depth interview. Although they may have been close to these people at the time, it is possible that their relationship dissolved and they do not desire further contact with these individuals. Elizabeth states, “There’s no relationship. Um, within the past week I’ve been getting threatening text messages, and threatening phone calls, so basically, I’m trying to avoid them.” This example highlights the transitory nature of some homeless youths’ social networks as former confidants can engage in harassing behaviors.

Social Network Members’ Housing Status

Housed

In addition to understanding the relationship status of each network member, we were also interested in learning where each network member resided because those who live on the streets may have more risk-enhancing characteristics compared to those who are housed. Interestingly, the majority of social network members listed were currently housed. When speaking about the housed members of their network, the homeless youth often noted that they were relieved at their housing situation. Elizabeth explains, “She lives with her mom in a very nice house that her mom can afford, and she always has food in her stomach, you know, and it, it’s a relief knowing that.” Elizabeth worries though that this same friend may become homeless in the near future and has warned her friend to continue living with her mom:

She’s 18 and hasn’t been homeless, but there’s a couple times where she’s been close [to being homeless] because her and her mom fight. And I just, you know….with the other people telling me, you know, how it is, and me knowing how it is, I’m like, ‘Don’t you dare! You got it good right now, you just stay there.’

Even though some of the homeless youths’ social network members may be currently “housed,” they were sometimes “doubling up” with friends (described below), which illustrates the marginal living situations of these individuals even though the respondent considered them housed.

Homeless

The next largest category of social network members included homeless individuals. While some of these network members lived on the streets, others stayed at shelters or led a nomadic existence moving from one friend’s place to the next. Lulu, a heterosexual American Indian, talked about the resourcefulness of one of her homeless network members. “She [is] kind of homeless but like she’d never like stay out in the streets. She always finds somewhere to stay.” Similarly, Stephanie reports that, “He is homeless. He does go to the shelter most nights, um, but he does have other living arrangements for weekends and certain nights of the week.” Although other homeless network members had limited housing options, they tended to be resourceful and were able to find a place to stay even if it was only for the night. Some youth described this process as “house-hopping:” “Well I think he’s house hopping now. I think he’s staying with Mitchell, another friend. But I-I’m not sure, usually I’m up to date with this stuff but I haven’t seen him in a couple days…” (Jennifer). Similarly, Nicole, a White heterosexual, reported how mobile their network members are: “He’s homeless…We’d bounce from hotel rooms every once in a while and he’s still homeless now.” Other homeless network members were not as lucky as these youth when it came to securing a comfortable play to stay for the night. Megan describes the situation of her network member who is her fiancé:

Um…R.J. [fiancé] currently either stays in T.J.’s [her friend’s] car, or she tries to sneak into her [T.J.’s] house. Um she doesn’t have a place to stay…I talk to her at like 9:30 at night, hoping to know where she’s gonna stay that night and usually she doesn’t know yet. Um so that’s difficult.

Other youth had network members who currently had a temporary residence as they were “couch surfing” or “doubling up” with friends. In these instances, homeless youth were staying with their friends even though they were not listed on the lease. These situations provide insight into the precarious nature of youth who are presumably housed. These homeless respondents are at high risk for returning to the street because they are currently staying with network members who were formerly homeless. Although many of the social network members do not routinely sleep on the street, they often lack consistent shelter and are in the same dire housing situation as our homeless respondents.

Other locations

Finally, a smaller number of social network members lived in foster care or out of town/state, were incarcerated, or the youth did not know of their network member’s housing location. Although the majority of homeless youth were able to list several people in their network, upon closer inspection there is evidence that some network members had more of a peripheral position in the network. In some cases, these individuals may have not been available on a daily basis to provide support. In contrast, other members were extremely important in the lives of homeless youth regardless of whether they lived on the streets or were housed. Our next section focuses specifically on the importance of network members in terms of the instrumental role they play in the lives of these homeless youth.

What Network Members Do for Homeless Youth

Although social network members may promote risky activities such as substance use and unsafe sexual behaviors (Ennett et al., 2006; Tyler, 2008), there are also risk reducing qualities about them including social support, protection from out-group victimization (Hagan & McCarthy, 1997), and the sense of belonging that comes with group membership (Ennew, 1994). Alternatively, other homeless youth report that their network members do not provide any assistance, which may be because they are unaware of the youth’s housing status or because they are unwilling or unable to provide aid.

Instrumental support

The social networks of homeless youth provided an array of functions on a daily basis such as instrumental support which includes furnishing shelter, food, money, and clothing. Given the dangers of sleeping on the streets, it is not surprising that securing shelter for the night is a main priority for homeless youth. Nicole explains, “He gives me places to stay and he will feed me and I’ve loaned him money before and it just kind of goes back and forth with whatever we need.” Lulu also reports that, “She’s my friend and she helps me out a lot whenever I need her and I help her out. Well she don’t really need my help, but like whenever she does, I’ll help her out whenever she needs [it].” Additionally, youth described how their network members told them about particular shelters in the area, which prevented them from having to sleep on the streets. This demonstrates the reciprocity that exists between some network members.

Social network members also provided the homeless youth with food, money, or clothing. Megan describes how her network member helps her and her fiancé (R.J.) out by supplying food and a place to sleep. According to Megan:

She [network member] helps me on the street because like, she would let me and R.J. sleep in her car if she couldn’t get us in her [network member] house, and she, um she works at a gas station so she like grabs the extra hot dogs at the end of the night that they would usually throw away so we’d have something to eat and brings us stuff to drink…

Melissa discussed how her network member not only provided her with clothing and food but also taught her how to secure food and other items for herself. This type of street socialization is common (cf. Hagan & McCarthy, 1997) and may offer the youth a sense of independence. Melissa explains:

Well [you] see, M.P., he takes care of me, he makes sure that I’m fed, and I have clothing and stuff. I mean he’s taught me how to go dumpster diving. You know? And um, supply for myself…to try to be independent…he showed me who’s your true friends and who aren’t and you can, you know, like tell the difference.

As such, social networks either directly provide food and other fundamental necessities or teach the homeless youth how to independently obtain these items.

Social support

Social support and advice were also extremely important to these youth. Network members also helped homeless youth by talking to them about their problems and encouraging them to remain sober or drug free. Michael explains: “She kind of fires, fires me in a way to go to AA meetings and stuff, ‘cuz’ she goes, so…it’s a good thing.” Additionally, Michael reports:

She is the type of person nowadays, um, that helps me stay away from my drug of choice, which is meth. She goes to a lot of meetings and she still believes we’re better than that and that’s a good thing…she inspires me…I guess something I do for her…is like, just be there for her…

Homeless youth also explain that this social support helps them live through their daily struggles and maintain a positive attitude. Elizabeth eloquently summarizes the role that social support by a network member plays in her life. “Um, K.J, she’s homeless like me, so we basically stand strong together. If you always have somebody by your side and somebody who knows what you’re going through, it’s easier to get through it.”

Some homeless youth have mental health issues and are unable to afford their medication or perhaps use it irregularly. Melissa for example discusses how her network member helps her to cope with anger:

Um, L.D., she helps me um, cope with my anger. ‘Cuz’ I am bipolar, and I haven’t…I’ve been off my medication for two years. My mom, well she’s kinda, actually, pretty proud of that because usually when I’m off my medication, I go all outa-whack, and I’ve been doing pretty good.

Having to cope with the social circumstance of being homeless, even among those without mental health problems, is extremely difficult and network members were there to calm youth and provide perspective on their situation. As such, homeless youth spoke about the importance of the support related to their well-being that they garnered from social network members. Stephanie talked about her network member and how they are there for one another, again attesting to the reciprocity that exists among these youth:

Um, we’re both each other’s anchors. He’s the one person who can keep me fully calm when I’m upset, and I’m the one person who can say the words that’s gonna help him calm down, ‘cuz’ he’s got a worse temper than I do.

Social network members also gave homeless youth advice on how to stay out of trouble, including avoiding conflict and not getting pregnant. Jennifer reports that her social network member “…keeps me in line. Like not starting fights and not being stupid and trying to do alright.” Stephanie also says that her friend gives her a lot of good advice: “For example, you know, just anything from sex and staying away from getting pregnant and that kind of thing to just your basic…where to go to when you need help on the street.” In addition to advice, their network members also worked to improve their self-confidence. According to Jennifer, a White bisexual, 19-year-old, “He [Vin] brings my self-esteem up. He tells me he has hope for me and that I can do a heck of a lot better than I have been doing in my life.” Additionally, when speaking about another network member [Doc], Jennifer reports: “He [says] …that I’ve survived through all this, so why not survive a couple more years.” The advice and hope that social network members offer to these youth likely plays a major role in their ability to live through each day and as such, are important for homeless youth’s well-being.

Protection

Social network members also provided respondents with protection, which often included watching out for the homeless youth so others would not take advantage of them. Stephanie states, “G.L. is the one that I’d go to for any sort of protection.” Nicole describes how her network member acts as a guardian: “…I guess like guardian kind of thing he looked out for me and stuff… Everybody knows him so nobody would mess with me or fuck with me because of him.” Females also discussed things that they did to reciprocate receiving protection. Nicole reports that, “[Be]cause we are both female, we usually got each other’s back when other people…all the guys try to chase us down at the library.” Nicole appears to suggest that there is safety in numbers and because of their similar circumstance, she and her network member watch out for each other. These findings are consistent with the literature as the social networks of homeless youth may provide protection from victimization on the street. Personal safety is a major concern for homeless people and being surrounded by social network members may reduce the risk of physical and sexual assaults by other homeless individuals. Despite this protection, homeless youth may still be victimized by those in their trusted social network (Smith, 2008).

Member does not help out

Unfortunately, other social network members do not assist homeless youth in their life on the street. According to Megan, network members do not always fulfill their promises. She recalls:

Um S.W. doesn’t do a lot of anything for me. I don’t do a lot of anything for him, and he doesn’t help me when I’m on the street; he actually hinders that. Um the first night that me and R.J. were on the street…we had talked to him [S.W.] earlier in the night telling him ‘we don’t have a place to stay [so] can we just come over for an hour or two to get warm?’ And um, we got over to his place, we knocked on the door, um he opened the door and the little chain thing was like locking it and he shut the door. And shut the lights off in his apartment. And it was just really hurtful…

Some youth reported that their network members were not aware that they were homeless which suggests that they may not have a close relationship and consequently glean little support from these network members. Additionally, social network members who had never been homeless may be unable to assist youth with day-to-day survival because they lack street knowledge and do not understand the homeless youth’s experiences. Although their housed peers could theoretically provide them with money or clothing, one respondent indicated that she did not think her housed network members could adequately anticipate her needs. Elizabeth believes that life on the streets is “real world” experience and is resentful that some of her network members have not faced these dire circumstances. She recalls:

I don’t think they have ever been homeless, so I think they haven’t really taught me anything about street smarts or anything like that, because of the fact that they’ve always had things handed to them. And I think that’s pretty sad. I think somebody just needs to kick them out and give them a taste of what real life is actually really about, you know.

Elizabeth’s quote further demonstrates that not all network members provide support and, as such, do nothing to help the homeless youth with their current situation.

What They Would Change about Their Network Members

Homeless youth discussed changing a number of things about their current network members such as personality characteristics (e.g., their attitude, emotions, how they listen), lifestyle (e.g., substance abuse, cheating), and their friends/partners. Some youth also indicated that they would not change anything about their network members. These themes provide more insight into the homeless youths’ social network dynamics.

Personality characteristics

Some homeless youth discussed the personalities of their network members and described these individuals as being very angry, which subsequently, made it difficult to get along with them. Jamal, a Black heterosexual male, mentioned that one of his network members has anger issues and he fears that she will eventually end up in a hospital after getting into a fight. Darnel, also discussed how he would like to change the personality of one of his network members because this woman does not appropriately cope with her anger as she takes out her frustration on others who are not even the source of the conflict. Two young women echoed these sentiments, reporting that although they got along well with their network members, these individuals were annoying at times and had attitude problems. In addition to anger, other personality characteristics the study youth wanted to change about their network members included their self-esteem. Amanda said that her network members’ low self-esteem made her upset. She says, “He’s a good looking guy, he’s kind of chubby…[but] he thinks he’s fat and worthless and it really makes me mad, because he believes what preppy girls tell him…they tell him ‘oh you’re ugly and oh you’re fat’ and he believes them and it makes me mad.” Finally, some youth said that their network member discounted other people and they wished they could change that. For example, Michael said that he would like to change the following about his network member:

Maybe the way she looks at the ghetto nowadays. Like, she used to live there and stuff and like now she’s like talks so much bad stuff about people when she sees them, ‘Oh my God is he still on Park Ave selling crack?’ That probably would change because she’s been there and done that and she knows how it is….You know what I mean? Like she never used to be like that, she never would really, would look at someone and like, look down on them and talk shit you know.

In summary, personality characteristics that homeless youth would like to change about some of their social network members include anger issues, attitudes, and levels of self-esteem.

Lifestyle

There were numerous types of behaviors and lifestyles that homeless youth wanted to change about their network members. Elizabeth for example, wished that her boyfriends were more faithful. She explains:

What I would change about him (J.E.) is for him not to be a cheater, because that’s why we broke up. And, J.D., there’s too much to say to change about him… [because] basically he’s a man-whore and he needs to stop; no, he needs to actually settle down with somebody and be there for somebody, you know, not just, you know, think of them as a booty-call.

Another common lifestyle theme that homeless youth wished to change about their network members included substance use. Nicole reported: “…he doesn’t need to get high anymore, he needs to sober up. When he’s sober he’s cool.” Other homeless youth described how a network member’s substance use is a reaction to a negative event. Megan for example explained: “But like, I see him a lot when he’s drunk, and I just, I don’t really like it. [Be]cause I know that he’s drinking because he doesn’t have custody of his son, and it’s just like a sad thing all together.” Another reason homeless youth wanted their members to stop using substances was because of the negative effect it had on the member’s health. Darnel explains:

…if there was some way that I could get into his head that, you know, the things that he’s doing…the way he like does drugs because…when I first met him he wasn’t into it as much as he is now. He looks like really bad; he’s only 23, no teeth-all false teeth. You know I feel like if there was like some way that I could talk to him and let him know, ’dude I used to be just as bad as you [and] you don’t have to continue that fucking way’ and it pisses me off because I felt like I more encouraged him to do it [use drugs] because when we first met each other…we were in jail for pretty much the same thing…

His quote indicates that he feels guilty for encouraging his friend’s substance use, especially since Darnel has been able to reduce his drug use whereas his friend is a frequent abuser. He struggles to find a way to help his friend quit using drugs but admits this process is challenging. Darnel explains: “Just to pretty much help him with like his drug problem and like find a way that I could talk to him more without sounding like I’m being a hypocrite because I used to do the same things that he did.” Our findings are consistent with Molina (2000) who reported that homeless men often acquire and circulate illegal drugs and alcohol among their social networks. Similar to Darnel, other homeless youth do not wish to judge their peers but want to do something to help. Michael explains:

Um, I guess we’ll always be friends no matter what, I mean I’m not going to judge her about- I mean I would really prefer her not to do meth, because I’m trying to stop, but, um, I’ve been clean for two months [be]cause I was in jail you know…but I’ve been doing pretty good. I really just really talk to her over the phone nowadays.

Distancing himself from his network member may make it easier for Michael to stay off of drugs.

Additionally, implicit in wanting their network members to change their lifestyle, homeless youth discussed how they wished these individuals did not have to grow up so fast, referring to the fact that they were teenage mothers and did not experience childhood. It is interesting that these homeless youth thought their peers were maturing too soon when they themselves are the same age and experiencing homelessness. Michael explains:

…I just feel like maybe she grew up too fast, [be]cause she’s only 18 and she acts so much more older, you know. Like, it’s basically kind of like she didn’t have a teen period basically. I didn’t either, but it’s- you know what I mean like we had to grow up…I don’t know how to explain it.

Michael continued his discussion of this network member and said, “Um, if I could change anything probably….just be about her having her, uh, her first baby when she was 16. Maybe let her grow up a little bit.” According to these homeless youth, certain lifestyle changes such as infidelity, substance use, and early entrance into adulthood would improve their relationships with their social network members.

Change their friends/partners

Some youth spoke about their network members’ legal trouble due to their associations with delinquent individuals and our respondents wished that they would end these harmful relationships. Megan described the consequences of her network members’ associations with other drug users:

I’d really like him to like get new friends. [That is] why he’s in jail; he just makes stupid decisions. And he’s very much a follower…And hopefully he won’t get max[imum sentence] for his charges, which would be like 55 years…And with that I think he could do better at not having these meth-head friends that get him into trouble and steal his stuff, and then he goes to jail for it. So yeah, it’s just his friends that kind of bother me but right now. It’s not really an issue because they don’t care enough to come visit him and call him and stuff.

Sarah is another youth who thinks her network member has made bad choices when it comes to choosing a partner. She vehemently states: “I can’t stand her husband. He is so annoying and full of crap…nobody likes him, he’s annoying and he cheats on her.” As such, some homeless youth would like to change their social network members’ associations.

No changes to network members

Despite all of the things that homeless youth would like to change about their social network members, there were a few who did not want to make any changes. Melissa indicated that she only interacts with those who appreciate her personal characteristics and that she feels the same about them. She explained: “I don’t want to change my friends. I like them for who they are. That’s why I like to hang out with them. That’s why they like me because they like me for me.” Finally, although a couple of youth said that there is nothing they currently would like to change about their network members, they were quick to add “yet!” (Michael).

Discussion and Conclusion

The narratives of this diverse group of homeless youth reveal that social networks are an important part of their life and they depend on these individuals for numerous functions including daily survival. Similar to the work of others (Ennett et al., 1999; Johnson et al., 2005), the social networks of homeless youth in our study tend to be heterogeneous, consisting of individuals from both home and the street. Our study also goes beyond descriptors of social network characteristics by describing initial network formation in detail. This is significant because some youth met their network members through shared circumstances of being homeless, which suggests that these members are more likely to be risk enhancing for homeless youth. In addition, knowing how networks are formed is important to service providers as they can tailor intervention strategies more effectively by knowing more about network composition and group dynamics.

We find that the composition of networks among our study youth tend to be varied in terms of gender, age, and role relationships. Social networks are often comprised of both males and females, which is different from the same-sex networks of general adolescent samples (Cotterell, 2007; Ennett & Bauman, 1994). We also find wide diversity in terms of age as the respondents had both young children and middle-aged adults in their networks, which is also unique from general adolescent populations. The role relationships that homeless youth have with their network members can be grouped into four categories: family, friends, partners, and other. When youth include a family member within their network, this person tends to be a brother or sister rather than a mother or father. This finding may be attributed to the fact that many homeless youth experience caretaker abuse and/or neglect (Tyler & Cauce, 2002; Whitbeck & Hoyt, 1999) so it is less likely that parents would be included in youths’ networks. Our respondents also mentioned that they have fictive kin in their networks who generally occupy a protective role. Friends are another common role relationship and they tend to provide support and advice.

Respondents also listed partners in their networks and some of these relationships were characterized by elements of risk. The term “partner” was often used loosely within this group. In some cases, the youth are steadily dating and committed to one another whereas in other situations, the relationship is casual and may only include sexual encounters or multiple partners. Discovering these nuances is significant because youth who have multiple sexual partners are at heightened risk for negative health outcomes including STDs (Greenblatt & Robertson 1993). Finally, “other” role relationships often include individuals who have more of a peripheral role in the network. For example, some respondents describe their network members as people with whom they consume alcohol and/or use drugs. This latter finding is consistent with existing literature, which shows that social network members promote risky activities such as substance use and unsafe sexual behaviors among homeless youth (Ennett et al., 2006; Tyler, 2008). However, it is important to reiterate that the social networks of these particular homeless youth are not entirely devoid of pro-social peers, a finding that is consistent with the work of Rice and colleagues (2007).

One recurring issue for our respondents is that not all of their network members fully understood their current circumstance. At times, youth felt resentful toward some of their network members who have never been homeless because they could not empathize with them. This has implications for homeless youths’ mental health and well-being and such ill feelings may lead to depressive symptoms and substance misuse. Respondents also have networks comprised of both homeless and housed individuals and each group brings unique dynamics to the network. On the one hand, having a housed member in one’s network may mitigate some risky behaviors because they may encourage homeless youth to engage in pro-social activities; however, housed members do not always understand homeless youths’ experiences and thus are not sympathetic toward them. On the other hand, having a homeless person in one’s network is positive because this person provides them with support and pertinent information regarding the locations of food and shelter. Despite this assistance, however, respondents noted that homeless network members tend to be risk-enhancing in terms of their involvement in unsafe sexual practices and substance use, which the words of the respondents clearly detail.

Homeless youths’ social networks also provide numerous functions such as instrumental support, which often includes providing access to or information on shelter, food, money, and clothing. Securing safe shelter for the night is a high priority for any homeless individual given the potential for being victimized while sleeping on the street. Reciprocity also exists among homeless individuals as they share money and other resources with one another when they could. Other network functions include offering social support, giving advice, and providing protection. Homeless youth are highly vulnerable so having peers who have “got each other’s back” (Nicole) is an important feature of their social networks.

Our final theme, which focuses on what homeless youth would like to change about their network members, is unique and adds to the existing literature because it reveals that many homeless youth often do not have much choice when it comes to network member selection, which is different from general adolescent populations. For example, some homeless youth indicate that they wish they could change the attitudes and the lifestyles of some of their social network members, providing insight into the lack of alternatives that some homeless youth have when it comes to selecting network members. Interestingly, a few youth report that they would not change anything about their network members. This is a response we would expect from general population youth who associate and spend time with their network group because they like them and enjoy their company. Although some homeless youth did not approve of the lifestyle or personality characteristics of their network members, they continue to associate with them, possibly because they have few other options or these members provide useful functions that are imperative for the youth’s survival.

In terms of limitations, our findings are cross-sectional, which do not allow for the dynamic nature of social networks and are not generalizable due to the reliance on a convenience sample. Although our qualitative sample included an overrepresentation of females, LGBT, young adults, and those residing in shelters, this was purposefully done given that these characteristics are potential sources of variability for our main focus of HIV risk behavior in the full sample. It is possible that oversampling on these personal characteristics may have influenced our findings as the inclusion of more males and heterosexuals may have changed our emergent themes.

Overall, the social networks of homeless youth tend to form through both conventional and unconventional means. Their composition tends to be heterogeneous in terms of the gender and age of the members and the role relationships that they occupy, which is unique from general adolescent populations. The distinctive blend of both housed and homeless individuals in these youths’ networks also provides some context regarding the complexity and social dynamics of their relationships and the potential problems this creates. Although housed youth may have resources to share, they do not understand the circumstances surrounding the homeless experience. Similarly, homeless network members are able to provide affinity but may create numerous stressors due to their lifestyle and/or personality characteristics. Although social network members provide a myriad of functions that benefit these homeless individuals, there are still things that some of our respondents would like to change about them. In conclusion, our study advances the literature on social networks and homeless youth by providing unique insight into the formation and composition of these networks and by revealing that in many cases, these young people do not choose their social networks but rather are chosen by their circumstance of being homeless.

At the policy level, our findings have implications for agencies and others who work with homeless youth. The heterogeneity we found among network members in terms of their backgrounds, role relationships, and behaviors poses challenges to intervention efforts that only target homeless individuals. Service providers should be aware of the dynamics of these groups when designing concerted intervention strategies. Our findings also revealed that many homeless youth still had ties to housed peers, which means that these individuals are not only sources of support for homeless youth but may also have risk reducing qualities that are beneficial to our participants. Programs are also needed that help build youths’ self-esteem, teach them about healthy relationships, and healthy coping styles. Without such programs, some youth will continue to turn to substance use as a way of coping with negative experiences such as custody issues with their children or early experiences of abuse and neglect. Finally, programs need to be tailored to meet the needs of different youth depending on early exposure to conflict and abuse, length of time they have been on the street, the types of risks they have experienced since leaving home, and the issues they currently face (e.g., substance abuse). The goal must be to provide youth with the tools and resources necessary to develop into healthy young adults.