Job Coaching and Success in Gaining and Sustaining Employment Among Homeless People

Abstract

Objective: People who are homeless experience many barriers that affect their ability to gain and sustain work. In this study, we investigate whether personal job coaching support contributes toward employment success. Methods: The short- and long-term employment outcomes of 2,480 clients participating in a labor market program were analyzed. Results: Clients being supported by a job coach have significantly higher chances of gaining employment than those not being supported. This holds particularly true for the youngest age-group. Furthermore, results also indicate that job coaching improves clients’ chances of successfully sustaining employment. Conclusions: Personal approaches and individual coaching seem to be promising strategies in social work practice and specifically in return to work programs for people who have experienced homelessness.

Keywords

homelessness labor market integration intervention job coaching

Introduction

In spite of efforts in social policy, homelessness in the United Kingdom and the Republic of Ireland remains a significant problem and one that has been intensified by the recent economic downturn. While in the period 2002–2009, the number of people in England accepted as homeless by Local Authorities declined sharply, the number of acceptances rose by 34% from 2009 to 2012. This increase has been attributed primarily to reforms in housing benefit and the effects of economic pressures on individuals and households (Fitzpatrick, Pawson, Bramley, Wilcox, & Watts, 2013). However, the number of statutory homelessness acceptances represents only a small proportion of the overall number of people who are homeless in England (Department for Communities and Local Government, 2013). Government statistics showed that in 2012, a total of 2,309 people slept rough in England on any one night (Department for Communities and Local Government, 2012b), a figure that represents a 31% increase from 2010 (Fitzpatrick et al., 2013). Other investigations estimate that up to 380,000 “hidden homeless” people are living in the United Kingdom (House of Commons ODPM, 2005). In the Republic of Ireland, the 2011 Census data recorded 3,808 people as either living in dedicated homeless accommodation or rough sleeping (Central Statistics Office, 2011). As is the case in the United Kingdom, it is likely that there is a high number of people who are “hidden homeless” in Ireland. In an assessment of housing need conducted in 2011, the Irish Government recorded that the number of people who were not reasonably able to meet the costs of their accommodation was 65,643, a 121.9% increase in figures from 2008, while the number of people who were sharing accommodation involuntarily was 8,543, up 71.9% from 2008 (Department of the Environment, Heritage & Local Government, 2008). Such dramatic increases are most likely due in part to the economic downturn, which resulted in higher unemployment levels in the Republic of Ireland than in the United Kingdom.

People who are homeless and those at risk of homelessness are among the groups of people needing intense support in preparing to integrate into the workforce. This is due to the fact that they often suffer from additional problems that act as barriers that reduce their ability to gain work, such as mental and physical health conditions, substance and/or alcohol misuse issues, or criminal convictions (Audhoe, Hoving, Sluiter, & Frings-Dresen, 2010; Muñoz, Reichenbach, & Hansen, 2005; Nordentoft & Wandall-Holm, 2003; Nusselder et al., 2013; Ramin & Svoboda, 2009; Wright & Tompkins, 2006). Employment has been considered a crucial step in ending homelessness, given its central “protective” role in peoples’ lives (Shaheen & Rio, 2007). Overall, estimates indicate that 77% of homeless people would like to work, yet only 15% currently were doing so (Singh, 2005). A study conducted in 2012 by homelessness agencies indicates that as few as 2–14% of people living in homeless hostels and supported housing were actually engaged in paid employment (Hough, Jones, & Rice, 2013). In labor market research, education and qualifications, as well as ethnicity and age, are among personal characteristics that are of importance for labor market participation and for sustained integration success of unemployed people (Gobillon, Rupert, & Wasmer, 2013; Koning & Raterink, 2013; Riddell & Song, 2011).

Labor market integration of socioeconomically disadvantaged homeless groups has become a challenge of high priority for social protection policies in the United Kingdom, the Republic of Ireland, and the European Union—recognizing the important role that employment plays in helping people to integrate into society and in promoting social inclusion (Department of the Environment, Heritage & Local Government, 2008; European Commission, 2013). The underlying principle governing welfare policy and reform is that work is the best route out of poverty. A fair and affordable benefit system and labor market inclusion are seen as crucial steps in efforts toward reducing poverty and welfare dependency (Department for Work and Pensions, 2014). In 2011, the cross Government Department Ministerial Working Group on Preventing and Tackling Homelessness declared helping people into work as one out of six aims toward reducing homelessness (Department for Communities and Local Government, 2011). In 2012, the Department for Communities and Local Government (2012a) highlighted “improving access to financial advice, skills and employment services” among its five commitments to preventing homelessness (p. 25). Labor market programs targeting people who are homeless have to cope with two major challenges. First, the target group is usually hard to reach and special emphasis is needed to outreach clients. Second, programs have to deal with clients who are highly deprived and are usually confronted with a variety of barriers against their integration into the labor market, as stated previously.

The “Ready for Work” Program

The Ready for Work program run by “Business in the Community” (BITC) aims to integrate people who are homeless or those at risk of homelessness into the labor market and is funded predominantly by the private sector. The program works with 155 businesses in 20 locations in the United Kingdom and Republic of Ireland providing training, work placements, and post-placement support and aims to equip people with the skills and confidence they need to gain and sustain employment. The choice of location for programs is determined by where there are the highest numbers of clients, of homelessness projects, and of employers. This means that programs are concentrated in large towns and cities rather than in rural locations. A prominent feature of this program is the degree of business involvement. It comprises an advisory group of senior business leaders who help steer the strategic direction of the program and the involvement of employee volunteers in delivering key elements of the program. The program has four stages:

Registration. Program managers meet prospective clients, referred by agencies such as homeless hostels, probation services, and other charities, to assess whether they are “work ready.”

Pre-placement training. Training takes place over 2 days to prepare clients for their placement and to build confidence to succeed in the workplace. This training is delivered by professional trainers and employee volunteers. Day 1 of the training is hosted in a community venue and Day 2 at a company.

Work placements. Companies provide 2-week unpaid work placements. Throughout the placement, clients are supported by a trained employee volunteer. Companies provide a written reference to help clients in their future job search.

Post-placement support. All program graduates are offered access to job coaches, job-seeking support, and further training. Companies provide employee volunteers to act as job coaches.

Each client has the opportunity to be matched to a job coach once he or she has completed their work placement. The job coach is a volunteer from a participating company who provides support and advice, helps with job applications, and who continues to strengthen their self-confidence and resilience. This program builds on nonspecialist volunteers with different backgrounds, personalities, and experiences. Job coaches and clients are matched, considering the needs of specific clients and the background and personality of both, job coaches and clients.

Job coaches participate in a 1-day training course, where they are instructed about coaching tools and where they receive information about the typical barriers to work their client might face. After this, they are matched with a client who has already completed a work placement. The aim is for the coach and the client to meet face to face on a weekly basis during the first 8 weeks. Fortnightly meetings then take place during the next 4 months. Typically, each meeting lasts an hour. However, both the frequency and the duration of the meetings are flexible and are subject to agreement between the job coach and their client. Meetings take place at the job coaches’ place of employment or another public place, for example, a cafe, or at the “Ready for Work club.” The content of job coaching meetings is determined by the client and their coach, but it may include job search activities and further preparations for specific job applications. Up to now, evaluation research on job coaching programs is scarce, with only weak evidence on respective intervention effects (Audhoe et al., 2010; Ferguson, Xie, & Glynn, 2012). This holds particularly true for job coaching interventions among homeless populations in the United Kingdom and the Republic of Ireland (White & Doust, 2011).

This study aims to fill this gap by investigating whether and to what extent being supported by a job coach contributes toward employment success of people who are homeless. In addition, we explore the potential impact of distinct socioeconomic factors on the probability of employment success. When analyzing employment success, we clearly distinguish two stages: first the probability of entering the labor market, that is, gaining employment, and second, the duration of remaining employed over a defined observation period. Both aspects are equally important in evaluating a possible intervention effect of the project described previously.

Method

Population and Data

This study is based on data collected continuously from January 1, 2009, to December 31, 2012. Employment outcomes were monitored up until August 7, 2013. The study population consists of homeless individuals participating in the Ready for Work program. Homelessness is defined according to the European typology on homelessness and housing exclusion classification that extends homelessness beyond rooflessness (rough sleeping or night shelters) to those living in insecure accommodation (e.g., “sofa surfing” or under threat of eviction) or inadequate accommodation (e.g., very overcrowded or unfit accommodation; European Commission, 2013). An adequate home is defined as having an appropriate dwelling (or space) over which a person and his or her family can exercise exclusive possession (physical domain), being able to maintain privacy and enjoy relations (social domain), and having a legal title to occupation (legal domain; Edgar, Harrison, Watson, & Busch-Geertsema, 2007).

All clients in the program were eligible to work in the United Kingdom or the Republic of Ireland, were 18 years or over when registering to the program, had been risk assessed by their referral agency if they had an unspent conviction, had expressed an interest in work, and had core basic skills, such as the ability to speak, read, and write English. All clients were referred to the program by a support worker from a homelessness or housing organization or a statutory service, for example, the Probation Service. Potential clients for the program were invited to a registration day to see whether Ready for Work was the right program for them. Following this, clients in conjunction with their support worker completed a registration form, submitted online or in paper format, which includes collecting biographical data.

Clients’ progress through the program and their employment outcomes are continuously monitored. BITC aims to keep in touch with its clients for at least 12 months, but clients may chose not to remain in contact. Twice a year a review of employment records was undertaken and if there has been no evidence of any contact within the last 3 months the employment record was ended.

Overall 4,402 clients participated in this program during the time period described previously. In this study, we included 2,480 clients with full data for analyzing the first research question. Seven hundred and forty-four of these clients successfully started employment and could therefore be included in the analysis of the second research question—the association between job coach support and the probability of sustaining employment. Five hundred and seventy-nine clients managed to sustain in employment for at least 3 months, 400 clients for at least 6 months and 210 for at least 12 months.

Statistical Analyses

In view of the fact that an experimental study design was not feasible for logistic and ethical reasons and that no randomization of the sample into those with or without job coaching could be done our statistical approach was restricted to exploratory analyses of available data. To this end, multivariate regression analysis was performed, taking into account the complexity of the data set and the need of adjusting for confounding effects. Our first research question, the association between job coaching and success in gaining employment, was analyzed by multivariate mixed logistic regression with random intercepts by region (Scotland, Wales, Republic of Ireland, and all nine regions of England¹). Respective analyses stratified by education, ethnicity, and age were adjusted for multiple testing. The second research question, the association between job coaching and success in sustaining employment, was analyzed by estimating a parametric survival regression model based on a Gompertz distribution for the hazard function. A random-effect intercept adjusting for regional variance was included after consideration of the Akaike information criteria and Bayesian information criteria. All respective statistical models were adjusted for a number of confounding factors. Factors included sex, year of terminating the program, ethnicity, length of unemployment (in five categories) prior to involvement in the program, having ever been alcohol dependent, having ever been substance abuse dependent, having ever been rough sleeper, age (in four categories), and education (higher level [National Qualification Framework (NQF) Level 3 or higher] and lower level [NQF Level 2 or lower]). All analyses were conducted with Stata 11. The coding and sample characteristics of all variables under study are presented in Table 1.

Table 1.

Sample Characteristics.

Variable	Categories	%	n	Percentage of Job Coaching Support
Gender	Male	70.28	1,743	24.21
Gender	Female	29.72	737	28.36
Age	18–24	27.62	685	20.15
	25–34	24.07	597	24.96
	35–44	18.31	454	29.07
	45–61	30.00	744	28.49
Education	Lower level	74.03	1,836	22.93
Education	Higher level	25.97	644	32.61
Ethnicity	White	64.80	1,607	19.10
	Black	25.73	638	26.15
	Asian	5.24	130	40.75
	Mixed race	4.23	105	28.57
Job coach	Yes	25.44	631
Job coach	No	74.56	1,849
Length of unemployment	<6 months	24.23	601	26.12
	> 6 months–1 year	22.86	567	26.98
	1–2 years	19.64	487	23.00
	2–4 years	16.69	414	25.36
	>4 years	16.57	411	25.30
Ever alcohol dependancy	Yes	12.50	310	21.29
Ever alcohol dependancy	No	87.50	2,170	26.04
Ever substance dependancy	Yes	19.84	492	18.50
Ever substance dependancy	No	80.16	1,988	27.16
Ever rough sleeper	Yes	27.90	692	24.13
Ever rough sleeper	No	72.10	1,788	25.95
Success in gaining work	Yes	30.08	744	43.70
Success in gaining work	No	69.92	1,734	17.59

Note. N = 2,480.

Results

The complete-case data set consists of 2,480 individuals of which 70% were men (Table 1). Six hundred and thirty-eight (25%) clients were supported by a job coach. The age of clients ranged from 18 to 61 years, and age categories were almost equally distributed across the sample. Sixty-five percent are White, and more than 50% have been unemployed for more than 1 year before the start of the program. Twenty-eight percent experienced a rough sleeping period, 13% have had an experience of alcohol dependency, and 20% an experience of substance abuse dependency.

Associations between job coaching and success in gaining employment are presented in Table 2. The chance of integration in the labor market is 3.70 times higher among those who were supported by a job coach as compared to those who were not supported by a job coach. This significant association between job coaching and labor market success in the overall group was further analyzed according to relevant socioeconomic characteristics in order to investigate the effectiveness of job coaching in different subgroups. In detail, we separated the analysis on the association between job coaching and success in gaining employment by age, ethnicity, and educational level. Correspondingly, in Figure 1, we report nine regression models with odds ratios (ORs) and adjusted 95% confidence intervals. The association between job coach support and success in labor market entry is significantly higher in young people aged 18–24 (OR: 7.44 [3.96, 13.98]) as compared to persons aged 45–61 (OR: 2.48 [1.53, 4.01]). Additional analysis with an interaction term in the pooled data (not shown) supports this notion. There is a continuous decline in the ORs across the age-groups. The second graph in Figure 1 shows the association between job coaching support and gaining employment in different ethnic groups. The ORs of job coaching are similar among all client groups, with the exception of Asian clients where job coaching seems to have a slightly greater effect. The regression models stratified by education show a slightly higher association of job coaching with employment success among those with higher education compared to those with no or lower level qualification.

Table 2.

Barriers for Gaining Employment.

Mixed Logistic Regression: Odds Ratios (95% CI)
Job Coach	Yes	3.70 [2.97, 4.61]
Age	18–24	Reference
	25–34	1.42 [1.07, 1.87]
	35–44	1.63 [1.21, 2.21]
	>45	1.55 [1.18, 2.03]
Education	High	1.15 [0.92, 1.43]
Unemployment length	< 6 months	Reference
	> 6 months–1 year	0.84 [0.65, 1.10]
	1–2 years	0.63 [0.48, 0.85]
	2–4 years	0.72 [0.53, 0.97]
	>4 years	0.66 [0.48, 0.90]
Ethnicity	White	Reference
	Black	0.88 [0.55, 1.42]
	Asian	1.20 [0.93, 1.54]
	Mixed race	0.98 [0.58, 1.63]
Alcohol	Yes	0.95 [0.70, 1.30]
Substance	Yes	0.87 [0.66, 1.14]
Rough sleeper	Yes	0.62 [0.49, 0.78]
Ll: −1,348.045
AIC: 2,736.09
BIC: 2,852.41

Note. N = 2,480; AIC = Akaike information criteria; BIC = Bayesian information criteria; Ll = Log likelihood. Adjusted for gender and year of finishing the program.

Figure 1.

Estimates of the associations between job coaching and gaining employment calculated by subgroups. Odds ratios with adjusted 95% confidence interval, adjusted for gender, year of finishing the program, unemployment length, ever been alcohol or substance dependent, ever been rough sleeper, age, ethnicity, and education, N = 2,480.

Returning to Table 2, the socioeconomic factors of age, length of unemployment, and whether an experience of rough sleeping seem to be important factors that affect an individual’s chances of gaining employment. Younger clients are less successful at getting employment than older ones. The length of unemployment before participating in this program seems to be another barrier against employment success. A long period of unemployment (more than 1 year) before the start of the program is associated with less chances of successful return to work if compared to a short period of unemployment (less than 6 months). Likewise, those with previous experience of rough sleeping have less chance of gaining a job. Yet, education, ethnicity, alcohol, or substance dependency, and gender (not shown) are factors that are not significantly associated with success in gaining work.

The employment outcomes of the 744 clients who gained employment were then used to study our second research question, the chances of sustaining employment. The associations between support by a job coach, socioeconomic factors, and the probability of sustaining employment are presented in Table 3. In this analysis, the probability of sustaining employment is analyzed in terms of hazard ratios of losing one’s job during the observation period. Values below 1 indicate longer “survival” in employment as compared to the reference group. Among these 744 clients almost half (43.7%) were recorded as being supported by a job coach following their work placement. Importantly, we observe again a significant association between job coach support and the probability of “survival” in employment. Clients being supported by a job coach are more likely to sustain employment than those who were not supported.

Table 3.

Barriers for Sustaining Employment.

Mixed Gompertz Regression: Hazard Ratios (95% confidence interval)
Job Coach	Yes	0.77 [0.64, 0.94]
Age	18–24	Reference
	25–34	0.73 [0.56, 0.95]
	35–44	0.77 [0.63, 0.95]
	>45	0.69 [0.54, 0.89]
Education	High	1.03 [0.85, 1.25]
Unemployment length	<6 months	Reference
	>6 months–1year	1.26 [1.02, 1.57]
	1–2 years	0.92 [0.71, 1.18]
	2–4 years	1.04 [0.79, 1.36]
	>4 years	1.17 [0.87, 1.57]
Ethnicity	White	Reference
	Black	0.80 [0.45, 1.39]
	Asian	1.00 [0.84, 1.19]
	Mixed race	0.68 [0.39, 1.16]
Alcohol	Yes	1.20 [0.88, 1.63]
Substance	Yes	0.85 [0.64, 1.13]
Rough sleeper	Yes	0.93 [0.76, 1.14]
Ll: −1,126.773
AIC: 2,275.547
BIC: 2,326.279

Note. Ll = Log likelihood. N = 744; adjusted for gender and year of finishing the program.

As is the case in the models for gaining employment, older clients have a higher chance of sustaining employment compared to the youngest age-group. However, length of unemployment, alcohol and substance abuse dependency, and an experience of rough sleeping are not significant factors influencing the chances of sustaining employment.

Discussion and Applications to Social Work

Within this study, two questions have been analyzed, first the importance of individual job coaching on gaining employment, and second the associations between job coaching and sustaining employment. Our analyses show significant associations between job coaching and success in gaining employment and with the chances of sustaining employment. This applies to clients of all ages but is most marked among younger clients (aged 18 through 24 years). The high effectiveness of job coaching among young people is important as our results suggest that in general younger clients perform less well on the program compared with those aged 25 and older. Finding positive interventions that help young people into work is key as there is now evidence that a period of unemployment while young can lead to permanent disadvantages over the life course (Roelfs, Shor, Davidson, & Schwartz, 2011; Scarpetta, Sonnet, & Manfredi, 2010).

Job coaching as a social work intervention includes many of the success factors identified by other studies including the use of individual training (rather than classroom approaches), engaging employers in design and delivery, building in support for transition to work, including support for job search whilst on the program, and personal support tailored to the needs of the individual (Brown & Koettl, 2012; Dench, Hillage, & Coare, 2006; Hasluck & Green, 2007; Wilson, 2013).

In trying to understand the reasons behind the significant relationship between job coaching and gaining and sustaining employment, it is useful to consider the reported experiences of the homeless clients who have completed the Ready for Work program and successfully gained employment. An associated qualitative study of a small sample of former Ready for Work clients who had successfully gained employment pointed out that job coaching was particularly useful by supporting navigating the jobs market and identifying jobs that match individual skills and experience. Furthermore, receiving support from job coaches may strengthen the clients’ motivation to continue job applications even in the light of recurrent failure (Ford, 2014). This study has several limitations. First, it is unclear to what extent the motivation of finding work is correlated with self-selection to take a job coach. The job coaching intervention is provided as an option in this particular program—the decision on making use of this assistance is up to the clients. Therefore, it is likely that a selection bias affected the association of job coaching and employment success. Additionally, the process of generating missing data is likely to be not completely at random. Second, features that are unique to the Ready for Work program itself, for example, privileged access to employers and access to the specialist knowledge and support of the Ready for Work manager following placement could affect the association between job coaching and employment success, particularly when combined with the effects of individual motivation (Ford, 2014). Third, skills of the job coaches and training of job coaches may impact a successful integration of the client (Vuori, Price, Mutanen, & Malmberg-Heimonen, 2005). Importantly, job coaches are volunteers from participating companies and receive only short training qualifications. The quality of support by the job coaches therefore possibly varied among different coaches but could not be controlled for in this study. Additionally, detailed information on job coaches (e.g., branch, type of employment, sociodemographic, and socioeconomic characteristics) was not available. Fourth, although our results suggest the existence of a positive relationship between job coaching and successful labor market entry, other factors beyond the scope of this study may impact a client’s chances of gaining and/or sustaining work—either positively or negatively. Such factors might include support from other support agencies who either are dedicated to helping the individual with other barriers to work, for example, a mental health condition or housing, or also provide support with looking for and applying for jobs, for example, probation services and other charities (Ford, 2014; Hough et al., 2013). Fifth, job-related factors might play a role, too. For example, the most frequent reason given by clients ending employment and recorded in the Ready for Work database is completion of a temporary employment contract. Therefore, the availability of suitable jobs might be an important factor. Fifth, we do not have data on the jobs of clients who successfully returned to work. Thus, the quality of employment (e.g., income and working hours) remains unclear.

In conclusion, results indicate that personal job coaching may improve the work ability of persons who are homeless. Our analyses suggest that job-coaching support is significantly associated with success in gaining and sustaining employment, especially among younger people. Personal job coaching support may be an important tool for social work practice and specifically for return to work programs supporting people who have experienced homelessness.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research leading to these results was done within the framework of the DRIVERS project () coordinated by EuroHealthNet, and has received funding from the European Community (FP7 2007-2013) under grant agreement no 278350. Johannes Siegrist was additionally supported by a Senior Professorship Grant from the Faculty of Medicine, University of Duesseldorf, Germany.

Notes

References

Audhoe

S. S.

Hoving

J. L.

Sluiter

J. K.

Frings-Dresen

M. H. W.

(2010). Vocational interventions for unemployed: Effects on work participation and mental distress. A systematic review. Journal of Occupational Rehabilitation, 20, 1–13.

Brown

A. J.

Koettl

(2012). Active labour market programms: Employment gain or fiscal drain (IZA Discussion Paper No. 6880). Institute for the Study of Labor, Bonn, Germany.

Central Statistics Office. (2011). Homeless persons in Ireland. a special census report 2011. Retrieved from http://www.cso.ie/en/media/csoie/census/documents/homelesspersonsinireland/Homeless,persons,in,Ireland,A,special,Census,report.pdf

Dench

Hillage

Coare

(2006). The impact of learning on unemployed, low-qualified adults: A systemic review (Research Report No 375). Department for Work and Pensions, London, England.

Department for Communities and Local Government. (2011). Vision to end rough sleeping: No second night out nationwide. London, England: Author.

Department for Communities and Local Government. (2012a). Making every contact count. A joint approach to preventing homelessness. London, England: Author.

Department for Communities and Local Government. (2012b). Rough sleeping statistics England—Autumn 2012 experimental statistics. Retrieved from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/284278/Rough_Sleeping_Statistics_England_-_Autumn_2012_-_REVISED.pdf

Department for Communities and Local Government. (2013). Statutory homelessness: July to September quarter 2013 England. Retrieved from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/264836/Statutory_Homelessness_3rd_Quarter__Jul_-_Sep__2013_England__2_.pdf

Department for Work and Pensions. (2014). Simplifying the welfare system and making sure work pays. Retrieved from https://www.gov.uk/government/policies/simplifying-the-welfare-system-and-making-sure-work-pays

10.

Department of the Environment, Heritage & Local Government. (2008). The way home: A strategy to address adult homelessness in Ireland 2008-2013. Dublin, Ireland: Author.

11.

Edgar

Harrison

Watson

Busch-Geertsema

(2007). Measurement of homelessness at European union level. Brussels, Belgium: European Commission.

12.

European Commission. (2013). Confronting homelessness in the European Union. Brussels, Belgium: Author.

13.

Ferguson

K. M.

Xie

Glynn

(2012). Adapting the individual placement and support model with homeless young adults. Child & Youth Care Forum, 41, 277–294. doi:10.1007/s10566-011-9163-5

14.

Fitzpatrick

Pawson

Bramley

Wilcox

Watts

(2013). The homelessness monitor: England 2013. London, England: Crisis.

15.

Ford

(2014). Interviews with ready for work clients (Unpublished report). London, England.

16.

Gobillon

Rupert

Wasmer

(2013). Ethnic unemployment rates and frictional markets (IZA Discussion Paper No. 7448). Institute for the Study of Labor, Bonn, Germany.

17.

Hasluck

Green

A. E.

(2007). What works for whom? A review of evidence and meta-analysis for the department of work and pensions (Research report no 407). Department for Work and Pensions, London, England.

18.

Hough

Jones

Rice

(2013). Longitudinal qualitative research on homeless people's experiences of starting and staying in work: Summary report. London, England: Broadway.

19.

House of Commons ODPM. (2005). Housing, planning, local government and the regions committee homelessness third report of session 2004-05, HC 61-I. London, England: House of Commons The Stationery Office Limited.

20.

Koning

Raterink

(2013). Re-employment rates of older unemployed workers: Ecomposing the effect of birth cohorts and policy changes. De Economist, 161, 331–348.

21.

Muñoz

J. P.

Reichenbach

Hansen

A. M.

(2005). Project employ: Engineering hope and breaking down barriers to homelessness. Work (Reading, Mass.), 25, 241–252.

22.

Nordentoft

Wandall-Holm

(2003). 10 year follow up study of mortality among users of hostels for homeless people in Copenhagen. British Medical Journal (Clinical research ed.), 327, 81. doi:10.1136/bmj.327.7406.81

23.

Nusselder

W. J.

Slockers

M. T.

Krol

Slockers

C. T.

Looman

C. W. N.

van Beeck

E. F.

(2013). Mortality and life expectancy in homeless men and women in Rotterdam: 2001-2010. PLoS One, 8, e73979. doi:10.1371/journal.pone.0073979

24.

Ramin

Svoboda

(2009). Health of the homeless and climate change. Journal of Urban Health, 86, 654–664. doi:10.1007/s11524-009-9354-7

25.

Riddell

W. C.

Song

(2011). The impact of education on unemployment incidence and re-employment success: Evidence from the U.S. labour market. Bonn, Germany: IZA Discussion Paper No. 5572. Institute for the Study of Labor.

26.

Roelfs

D. J.

Shor

Davidson

K. W.

Schwartz

J. E.

(2011). Losing life and livelihood: A systematic review and meta-analysis of unemployment and all-cause mortality. Social Science & Medicine, 72, 840–854. doi:10.1016/j.socscimed.2011.01.005

27.

Scarpetta

Sonnet

Manfredi

(2010). Rising youth unemployment during the crisis: How to prevent negative long-term consequences on a generation. Paris, France: OECD Publishing.

28.

Shaheen

Rio

(2007). Recognizing work as a priority in preventing or ending homelessness. The journal of primary prevention, 28, 341–358. doi:10.1007/s10935-007-0097-5

29.

Singh

(2005). No home no job. Off the streets and into work. London, England: Crisis.

30.

Vuori

Price

R. H.

Mutanen

Malmberg-Heimonen

(2005). Effective group training techniques in job-search training. Journal of Occupational Health Psychology, 10, 261–275. doi:10.1037/1076-8998.10.3.261

31.

White

Doust

(2011). Coaching into employment. Evaluation of the in work staying better off programme. London, England: Crisis.

32.

Wilson

(2013). Youth unemployement: review of training for young people with low qualifications. London, England: BIS Research Paper number 101. Department for Business Innovation and Skills.

33.

Wright

N. M.

Tompkins

C. N.

(2006). How can health services effectively meet the health needs of homeless people? The British Journal of General Practice, 56, 286–293.