Escaping Homelessness 1 Running head: ESCAPING HOMELESSNESS Escaping Homelessness: Anticipated and Perceived Facilitators

One study with two distinct sections was conducted to identify factors facilitating escape from homelessness. In section one, 58 homeless individuals rated possible facilitators of escape (factors they believed would help them become more independent and self-sufficient). In section two, 80 participants who had already exited homelessness rated the same facilitators (factors that would have helped them become more independent and self-sufficient) and the importance of actual factors that facilitated escape. When rating factors in the hypothetical, both groups rated obtaining housing as being particularly important for facilitating movement toward independence. People formerly homeless who reported perceived facilitators of escape, however, also reported that their escape was facilitated by realization of their own abilities and potential to offer something to the world. The findings have implications for the design of community interventions helping individuals who are homeless. Escaping Homelessness 3 Escaping Homelessness: Anticipated and Perceived Facilitators Many aspects of homelessness deserve the attention of academics. This study, in particular, focused on facilitators of escape from homelessness as perceived by people currently or previously homeless. Understanding the perceptions of people who are homeless is important because an understanding of their perceptions can improve social policy. Policy that ignores the felt needs of people who are homeless may not only lack compassion for these vulnerable people, but may also fail to achieve objective goals such as helping people move towards independence. An extreme interpretation of Maslow’s (1943; 1948) hierarchy of needs might suggest that people who are homeless will give priority to lower level needs such as safety, food, and shelter, rather than higher level needs such as social and esteem needs. However, some prior studies suggest that higher level social and esteem needs (e.g., “realizing one’s self-worth;” MacKnee & Mervyn, 2002) may play an important role in helping people escape homelessness. The rating scales used in the current study can help clarify the importance of these different constructs from the perspectives of people currently or recently homeless. Pathways Out of Homelessness Prior research on factors perceived to facilitate escape from homelessness have been mainly qualitative in nature (e.g., Kidd & Davidson, 2007; MacKnee & Mervyn, 2002; MorrellBellai, Goering, & Boydell 2000; Raleigh-Duroff, 2004). A careful reading of these prior studies suggests some commonalities in the findings and posits a smaller number of underlying constructs which may explain the events facilitating escape from homelessness. These constructs include some that might seem obvious such as 1) access to housing and economic stability, 2) events facilitating control of substance abuse, and 3) events facilitating treatment for health problems (including mental health problems), but also others that might seem less obvious such as 4) events providing social support, 5) events facilitating recognition of self worth and ability, 6) events related to managing past and present issues and responsibilities, 7) events facilitating recognition of the negative aspects of the street, and 8) spirituality. Further research related to each of these clusters will be reviewed. Facilitator #1: Access to Housing and Economic Stability. Some researchers argue that homelessness is a temporary state that can be resolved through the provision of housing (Goodman, Saxe, & Harvey, 1991; Shinn, 1997). This approach has some important merits and is supported by research which has found a link between housing cost and homelessness (Lee, Price-Spratlen, & Kanan, 2003). Economic stability is of course necessary to retain housing (Ratcliff et al., 1996). Job skills training, one means to economic stability, has been cited by homeless individuals as being important in helping them escape the street (Ratcliff, Shillito, & Poppe, 1996). Employment has been supported as an important factor in contributing to overall well-being across the general population (Huppert, & Whittington, 2003). Thus, we expected that previously homeless individuals would cite factors related to economic stability as being helpful. Nonetheless, while housing does appear to play a significant role in escaping homelessness, it should not be assumed that housing alone is a sufficient pathway to independence for all people who are homeless. Some researchers have found evidence that the problem of homelessness frequently extends beyond housing. Wolf, Burman, Koegel, Sullivan, and Morton (2001) examined changes in subjective quality of life among homeless adults who obtained housing. The researchers investigated those who did not exit homelessness, those who found dependent housing, and those who found independent housing. Those who obtained independent housing exhibited the largest increase in overall quality of life. However, even for this group, the overall increase in life satisfaction was small (see also Wolf et al., 2001) suggesting that the needs of many of the participants extended beyond mere housing. Escaping Homelessness 4 Facilitator #2: Events Facilitating Control of Substance-Use. Some prior research and theory suggests that events facilitating control of substance abuse will facilitate escape from homelessness (Zlotnick, Tam, & Robertson, 2003). As such, we expect that previously homeless individuals with substance-use disorders would tend to rate events related to substance-use as most important in the transition off the street. Facilitator #3: Access to Health Care including Mental Health Care. The high rates of mental health problems among people who are homeless (BC Ministry of Social Development and Economic Security, 2001; Gelberg & Linn, 1989) suggest that mental health care may facilitate escape from homelessness. While diverse mental health issues surround homelessness, depression is particularly common (Wong, 2000). Escaping homelessness can take much perseverance. One must negotiate the challenges of bureaucracies, the challenges of finding work while lacking a fixed address, and the challenges of maintaining sobriety when surrounded by opportunities to abuse substances. Depression may frequently rob people of the motivation necessary to maintain this type of perseverance. Facilitator #4: Access to Social Support. Escaping the street often requires instrumental support from others (e.g., material goods, transportation; Macknee & Mervyn, 2002). Emotional support may also improve well-being and provide resilience in the face of life stressors (Cohen & Willis, 1985; Muller, Goh, Lemieux, & Fish, 2000). Bassuk, Mickelson, Bissell, and Perloff (2002) have provided evidence that positive social support is associated with improved psychological well-being following a traumatic event. Those who lack social support may be overcome with stress and thereby fail to accomplish the tasks necessary to escape the street. However, past research has found that lower socioeconomic status is related to less social support (Shinn, Knickman, & Weitzman, 1991). To date, there has been limited research on the role of social support among those experiencing extreme poverty (Bassuk et al., 2002). Facilitator # 5: Events Provoking Recognition of Self Worth and Ability. Research by Schweitzer, Hier, and Terry (1994) has shown that homeless adolescents report significantly lower self-esteem than do most adolescents. Self-esteem has been found by Diblasio and Belcher (1993) to be related to goal attainment among the homeless. Also, both Macknee and Mervyn’s (2002) research and Diblasio and Belcher’s (1993) qualitative research suggest that events increasing self worth could be an important facilitator for escaping homelessness. To our knowledge, no other studies prior to the current study have asked people who have already escaped homelessness to rate the importance of recognizing their worth in their efforts to escape the street. Facilitator #6: Dealing with Past and Present Responsibilities (Excluding Substance Use). There was reason to believe that some participants would report that dealing with past and present responsibilities helped them escape homelessness. MacKnee and Mervyn (2002) developed this construct based on their interviews with formerly homeless youth during which participants cited three factors related to responsibilities. These included taking responsibility for the issues that caused them to become homeless, striving to fulfill the responsibility of being a parent, and beginning to feel accountable for previous decisions and present conditions. One could argue that substance use issues deserve to be included in this construct, however, it could also be argued that substance use issues are sufficiently frequent that they may deserve their own construct. Furthermore, some individuals may deal with their substance use yet not feel a need to deal with other past and present responsibilities. Facilitator # 7: Events Provoking Intensified Cognizance of the Negatives of the Street. There is also evidence to suggest that motivation to change is much stronger if the person has both a detailed vision of a hoped for positive state they are seeking and perceives negatives of the Escaping Homelessness 5 current state (Cox, 1998, Cox & Blount, 1998, Cox & Kilinger, 1988). Thus, Cox and Kilinger (1988) argued that in order to facilitate personal change, an intervener may need to bolster motivation to leave the current state. Based on this theoretical orientation, and the research of Kidd and Davidson (2007), we expected that events enhancing negative perceptions of the street would frequently be perceived as significant in escaping homelessness. Facilitator #8: Religious or Spiritual Experiences. MacKnee and Mervyn (2002) found that a subset of their participants reported that spiritual experiences enabled them to escape homelessness. Though many services to people who are homeless are operated by religious organizations, we know of no study that has interviewed previously homeless participants to learn how many of them attribute their life change to religious or spiritual experiences. Current Research While the aforementioned studies and the suggested constructs provide some insight into factors enabling escape from the street, there is still much to be explored. The current study is a cross-sectional analysis of people who were currently and previously homeless to assess the perceived importance of these types of factors in facilitating escape from homelessness. Past research on exits from homelessness has been mainly qualitative in nature, the current research however, built on the important past research through the use of quantitative methods. Though qualitative research is valuable, quantitative research can extend those prior findings with a larger sample and allow a more systematic assessment of the importance of each facilitative factor. The use of rating scales in the current study also allowed both theoretical considerations and basic item analytic techniques to be used to reduce the number of constructs to those that are most central. The first part of the current study consisted of interviews with individuals who were currently homeless during which they were asked about factors that would likely help them escape homelessness (anticipated factors). In the second part, people who had already escaped homelessness were asked the same hypothetical questions (factors that would have helped) and also asked about perceived events that actually facilitated their escape from homelessness. The purpose of obtaining information from both the currently and previously homeless was to develop a comprehensive set of factors which summarize the anticipated and perceived needs for escaping homelessness of those who have and have not escaped street life. Section 1 In section 1, participants who were currently homeless rated the extent to which they needed various supports in order to move toward greater independence and self-sufficiency. Hereafter, these will be referred to as anticipated factors as participants were rating the extent to which these supports were needed, even though the participants may not have had access to these supports. Method Participants. Fifty-eight participants were recruited at a homeless shelter in Vancouver British Columbia through the use of flyers posted in the foyer and elevators. Participants saw the sign and contacted the researcher to arrange a meeting. The interview location is both a 70 bed emergency shelter and a 120 bed supported housing facility. The facility provides the basics of subsistence, counseling and referral services, education and skills development, pro-bono legal services, and recreational activities. While data from the currently homeless were collected from only one location, the transient nature of homelessness (Cloke, Milbourne, & Widdowfield, 2003) improves generalizability as we were likely sampling individuals from a variety of different backgrounds. Participants were required to be fluent in English and meet the criteria for being considered homeless. Homelessness was defined as (a) staying outdoors from evening until Escaping Homelessness 6 morning because they lacked acceptable housing and/or (b) staying in an emergency shelter because they lacked acceptable housing. Each participant was provided with a $25 gift certificate at the start of the interview and told that they could quit at any time. There were 44 men (average age of 44.6, SD = 9.18) and 14 women (average age of 40.5, SD = 11.90). This ratio is generally representative of the demographic makeup of the shelter. Forty five percent of participants reported that they had been diagnosed with a mental illness, and 57% reported that they had taken medication for a mental illness. A significant proportion reported a current or past drinking problem (66%) or drug problem (64%). The questions used here were part of a larger interview which asked questions about well-being and social networks. Measures. Participants were asked to rate the extent to which they needed each of 25 different supports in order to move towards greater independence and self-sufficiency. Items were developed by conducting a series of focus groups with people recently homeless, by consulting with people working in temporary shelters, and by reading published research studies. Each item began with the following stem, “In order to become more independent and selfsufficient you need to be provided with ____.” The endings included items such as “housing,” “food,” etc. Participant ratings ranged from 1 = strongly disagree to 7 = strongly agree. Results In order to suggest a smaller number of underlying constructs, an oblimin rotated principal component analysis was conducted. This procedure groups together items that tend to be treated as similar by the participants and helps identify a smaller number of latent variables. The analysis suggested grouping the items into three categories: social support (e.g., having a mentor, seeing a counselor, receiving training in how to handle stress), services (e.g., transportation, dental care, and social service funding), and a third component including food, housing, and negatively weighted substance use treatment. We then calculated average scores for each group of items with some exceptions. Items assessing housing, health care, food, and treatment for substance use were separated because much effort and spending goes into each of these domains, so it seemed important to separate participant ratings of need for these types of services. The mean rating for each is displayed in Table 1. The reliability of the two remaining scales was acceptable (alpha for services = .76; alpha for social support = .94). Housing toped the list of factors perceived to be most necessary for movement towards independence followed by health care, food, services, support, and treatment for substance use. In order to provide more detail for the reader, Table 2 shows the average ratings for each item from the anticipated questionnaire. At the item level, participants again rated the value of housing, health care, and food as being particularly important in leading them towards independence. Discussion The findings suggest that most of the participants perceived housing to be their greatest need. These results suggest that our participants agreed with a housing first type of approach. Many participants also rated medical care as important for facilitating their escape from homelessness. The health item used in the first part of the current study did not differentiate between general healthcare and mental healthcare. Therefore, in the second half of the study, people recently homeless were specifically asked about the role of mental health care in their escape from homelessness. One of the more surprising findings was the relatively low rating given to treatment for substance use. This finding highlights a fact easily forgotten: People who are homeless are not homogenous; some perceive a need for treatment of substance use, while others may not perceive such a need. The reasons for this disinterest in treatment will probably Escaping Homelessness 7 vary: Some may distrust the efficacy of treatment. Others may not want to end their substance abuse. Others will not need treatment because they do not abuse substances. Section 2 The second half of the study provides further information about facilitators of escape from homelessness. In section two, people who had successfully escaped homelessness completed questionnaires exploring their escape. In order to allow comparison with the currently homeless, the same questionnaire on anticipated supports was again included in the study. This questionnaire asked about anticipated factors that would have facilitated movement towards independence when they were homeless. However, the second half of the study also used a more theory driven approach. In particular, a new questionnaire was constructed that included enough items to create subscales representing each of the eight types of supports found in prior qualitative studies and reviewed in the introduction to this paper. This new questionnaire asked about events that had actually enabled the participants’ escape from homelessness (actual facilitative events rather than hypothetical supports). Method Participants. Participants were recruited through the use of flyers posted at low-income housing apartments and community centers. Interested participants telephoned or emailed the researcher to arrange a meeting time. During the initial conversation participants were screened for the following prerequisites. To participate, individuals were required to be at least 19 years of age, fluent in English, previously homeless for a period of at least one month, and off the street in stable housing for at least six months. Homelessness was defined as (a) staying outdoors from evening until morning because they lacked acceptable housing and/or (b) staying in an emergency shelter because they lacked acceptable housing. Interviews took place in both Surrey and Vancouver, British Columbia. In Vancouver, interviews were performed at a drop in centre frequented by the previously homeless. This center is available to everyone and there are no barriers to its use. In Surrey, the interviews were conducted at a 30 bed supported living facility for both men and women. Individuals are able to live in this facility for up to two years. Clients are expected to live independently (e.g., managing finances, meals, healthcare), but with a reduced rate in rent. The male to female ratio at both the supported living facility and the drop in centre reflect the ratio we obtained in the current study. Eighty individuals participated with a range of ages between 21 and 62 (average age = 42.4, SD=10.36). The sample consisted of 61 men (average age = 43.6, SD=10.27) and 19 women (average age = 38.8, SD=10.04). Of the 80 participants surveyed, 51 indicated that when they were homeless they regularly used illegal drugs other than marijuana. Thirty-eight individuals indicated that they had a history of mental illness. The vast majority of those reporting a mental illness reported depression, although, bi-polar disorder and anxiety disorders were also commonly cited. The mean for the longest period of homelessness was 21.69 months (SD = 28.88) and the mean amount of time off the street at the time of the interview was 25.05 months (SD = 35.81). Finally, 58 individuals reported that they currently lived in a single room occupancy unit. The average reported monthly cost of an apartment was $361.48 (SD = $117.13). Measure of Anticipated Facilitators. The questionnaire included the same anticipated facilitator questionnaire given in Section one (e.g., asking whether each factor would have facilitated movement toward independence at the time they had been homeless). Measure of Perceived Facilitators. Participants were also asked to respond to 31 items assessing perceived events that facilitated their movement out of homelessness. To our knowledge there is currently no scale which assesses facilitators of escape from street life. Thus, we undertook the task of developing a scale to assess the components which underlie exits from Escaping Homelessness 8 homelessness. With the use of previous literature, highlighted in the introduction of this paper, we developed a 31 item measure. The questions were created from the literature reviewed and especially based on the work by MacKnee and Mervyn (2002), but also relying on the work of Raleigh-Duroff (2004) and Morrell-Bellai et al. (2000). Specifically, participants were asked, “In your experience, how did each of the following help you in your efforts to escape homelessness?” The questionnaire had 6 response options ranging from very harmful (-2) to very helpful (2) or did not happen (0). Table 3 includes the means and SD for all 31 questions. We theorized that the items in the perceived events questionnaire could be grouped into scales according to the constructs reviewed in the introduction to this paper. For example, people who found one form of economic support to be helpful would also tend to find other forms of economic support helpful, thus, the economic support items could be clustered into a single scale. We expected likewise for most of the other constructs. One could argue that for items asking about actual events rather than attitudes, internal consistency could be expected to be low because potentially important items could not be rated by people for whom the event had not occurred. Thus, when calculating internal consistency (Cronbach’s alpha) for each construct, we excluded people who could not rate an event because it did not happen to them. The reliability indices were as follows: social support (alpha = .65), substance use (alpha = .74), dealing with the issues that caused homelessness (alpha = .73), realizing the negatives of homelessness (alpha = .61), realizing self-worth (alpha = .62), and economics (employment and skills training) (alpha = .80). There were 3 individual items that were not clustered for theoretical reasons including spirituality, treatment for mental illness, and receiving housing. Results Anticipated Facilitators. The ratings for the anticipated facilitator scales (factors they believed would have helped them) are displayed in Table 1. These scores were calculated exactly as they were in Section 1. The participants who had escaped homelessness ranked housing as most important and food as the second most important factor that would have enabled movement towards independence. Next, they affirmed the importance of health care, then services (transportation, dental care, and social service funding), followed by social support (e.g., having a mentor, seeing a counselor, receiving training in how to handle stress, and how deal with government workers), and finally treatment for substance use. These results appear remarkably similar to those for the currently homeless in Section 1. In order to determine whether the currently and previously homeless agree on the ranking of what would be (or would have been) helpful, the degree of agreement or concordance between the different sets of ranks were calculated using Kendall’s coefficient of concordance. The value of Kendall’s coefficient was W = .935, p = .006. Thus, both groups appear to generally agree on the order of importance of the anticipated helps. Both groups appear to agree with the housing first philosophy. For completeness, Table 2 provides the entire item level results. Perceived Events. The previously homeless were also asked to rate the significance of events that facilitated their escape from homelessness. One purpose of the analysis was to reduce the large number of events found by MacKnee and Mervyn (2000) into a smaller number of underlying clusters of event types that facilitate escape from homelessness. Clusters were created based on the theoretical categories discussed above. The results showed that housing was perceived as the single most important event that had assisted their escape from homelessness followed by realizing self-worth, realizing the negatives of the street, social support, dealing with past and present issues and responsibilities, spiritual awakening, mental health treatment, substance-use issues, and finally economics (See Table 4). A complete item list is presented in Table 3. Escaping Homelessness 9 Discussion When rating the anticipated factors facilitating escape from the street, the currently homeless responded much like the previously homeless. Both groups gave priority to items meeting basic physiological needs: Shelter, food, and health care. This finding affirms much of the work of organizations that seek to support people who are homeless by providing for these essential needs. However, a somewhat different pattern emerged when formerly homeless individuals rated perceived events facilitating escape. Though housing was again viewed as the most important event facilitating escape from homelessness, realizing self-worth was perceived as the second most important event by the sample. Realizing the negatives of the street was a close third. Social support, dealing with past and present issues and responsibilities, having a spiritual experience, and receiving treatment for a mental disorder followed closely. Interestingly, dealing with substance-use issues and obtaining employment and/or completing a certificate or program at school were last in terms of helpfulness for escaping homelessness. It is somewhat surprising that substance abuse treatments received a relatively low rating in terms of facilitating escape. One possible explanation for the low mean for substance-use is that the questions that composed the cluster did not correctly target the events that tend to promote an end to substance-use (e.g., Question #30: “Experiencing a drug overdose” may not tend to promote an end to substance use). Another possibility is that substance abuse treatment is very important for some individuals while others may see their substance use as a secondary concern bringing down the mean rating for the group as a whole. General Discussion The focus of the current research was to determine what was/would be helpful in escaping homelessness as reported by people who had experienced homelessness. The findings can provide guidance regarding both theoretical and policy questions. From a theoretical perspective, it is interesting to ask questions such as whether movement off the street is best facilitated by focusing on food, shelter, medical care, and other physiological needs (lower needs on Maslow’s, 1943, hierarchy) or also by directing some resources toward social needs, self-actualization needs, and other needs. The results suggest that peoples’ reported answers show some consistency but also depend, in some ways, on when and to whom you ask the question. For example, both currently and recently homeless people identified housing as particularly important for escaping from homelessness. Health care and food were also rated as important. This response pattern supports the idea that subsistence level needs are perceived as very important by people who are homeless and by people who have been homeless. In some ways, this finding is unsurprising. However, when people who had already escaped homelessness were asked about events that they perceived to actually facilitate their escape (rather than rating events that would have helped them escape), an interesting finding emerged. Specifically, internal cognitive events (e.g., realizing your self-worth) were rated as having been important. Of course, this finding does not suggest that fewer resources be devoted to meeting subsistence needs, but it does suggest that the role of internal cognitions should not be underestimated. Social, or even therapeutic interventions, may play a very significant role in promoting these types of cognitions. It may seem somewhat surprising that these results put so much emphasis on internal cognitions rather than solely emphasizing external supports. Attribution research in social psychology, especially, the frequently cited actor-observer effect (Jones & Nisbett, 1987), suggests that outsiders viewing people in the midst of tragedy may often blame the victim and Escaping Homelessness 10 assume that the victim needs to take responsibility to improve their lives. In contrast, the same actor-observer effect would suggest that people in the midst of tragedy will be more likely to blame their circumstances and, as a result, believe that outside support will be the pathway out. However, the people who were formerly homeless give a more sophisticated response, attributing their escape to especially external supports (housing, food, and healthcare), but also to internal cognitions. Practical Application We suggest caution in interpreting these results because people, whether homeless or not, may misperceive some of their own needs. Nonetheless, the results deserve attention. The results, especially for people who have escaped homelessness, suggest that resources be focused on housing, other basic physiological needs, and helping individuals realize their self-worth. Consistent with Maslow’s hierarchy of needs, housing and nourishment appear to be of foremost importance in escaping homelessness. Without meeting one’s basic needs it is difficult to move on to dealing with more complex issues. Second, there appears to be a need to increase confidence and self-worth among the homeless. Some debate has emerged within psychology as to whether raising self worth is best done by social affirmation (e.g., shelter staff saying, “I believe you have a lot of potential”) or instead through training in new skills so that people will see irrefutable evidence of their own potential. Either way, building an individual’s sense of competence and efficiency is likely very useful in helping the homeless escape the street. Limitations While the current research is very valuable, there are a few limitations. First, the small sample size has a negative influence on our ability to draw firm conclusions from the principle components analysis. Further, subjects were identified and used based on their willingness to participate in the study. Participants who volunteer to talk about their experiences with homelessness likely differ from those who do not volunteer thus, limiting generalizability. The generalizability of the current research is also limited by the physical location of participants; all were recruited from the lower mainland, British Columbia, Canada. Admittedly, the high coefficient of concordance does suggest at least good cross-sample generalizability (i.e., two quite different samples produced quite similar results), thus providing at least some evidence that the results may generalize further. Nonetheless, a larger sample from a more diverse geographical area would allow for more reliable generalizations to be drawn. Future Directions As an extension to this study it would be very interesting, and useful, to interview those individuals who have attempted yet failed to escape the street. Comparing the responses from those who have and have not escaped homelessness may further highlight the events vital to escaping the street. Homelessness is a major problem in North America. However, pathways out of homelessness have not been sufficiently investigated. Escaping Homelessness 11