Challenges for shelters for abused women in Taiwan

Abstract

This study explored the challenges faced by shelter workers. A qualitative survey was tape-recorded and featured 19 focus groups comprising 6 representatives of the Department of Protective Services, 43 local government representatives and 84 shelter workers from 26 women’s shelters in Taiwan. The verbatim transcriptions were subsequently analysed. Finally, two themes were identified: resource challenges and regulations governing shelters. Understanding shelter workers’ voices is key to ensuring that the authorities and the public pay attention to the challenges faced by shelters. In addition, future implications were discussed.

Keywords

Challenges operation shelter workers Taiwan women’s shelter

Introduction

Taiwan’s central and local governments continue to develop policy initiatives to address domestic violence (Ministry of the Interior, 2017). Among initiatives designed to help victims of abuse, those that focus on shelter and housing are particularly vital. While some studies have indicated that Taiwanese abused women sought help from criminal justice system, family and friends, religion, domestic violence shelters and medical assistance (Shen, 2009), a recent study determined that emergency shelters and housing assistance were among the most accessed types of victims’ services in Taiwan (Yu, 2015). This shows that women’s shelters offer important assistance when abused women need a place to stay.

According to Taiwan’s Ministry of Health and Welfare (2017), 64 shelters (the term ‘shelter’ is used broadly to refer to all residential facilities for all types of victims, including 26 abused women’s shelters) operated across Taiwan in 2016. This was an increase of 22 shelters since 2004 (an increase of 34.38%) and 6 shelters since 2014 (an increase of 9.4%). Since 2008 the number of shelters in operation has increased or remained stable, according to all central and local governments for which data are available.

Abused women staying in women’s shelters are a prominent subgroup of domestic violence victims, reporting relatively severe abuse (Arnold and Ake, 2013; Bybee and Sullivan, 2005), posttraumatic stress disorder symptoms (Jarvis et al., 2005), and related injuries (Wang, 2006). Although the principal goal of shelters is to offer safety, the places in which abused women stay may also represent a unique opportunity to address abusive relationships and problems that are either a cause or a consequence of abuse (Yu, 2013). Indeed, women’s shelters strive to improve women’s internal resources, such as their self-esteem and self-efficacy (Huang, 2015). Shelters offer multiple services to abused women, including support, advocacy, resources for children, social service networking, mental health referrals, legal and financial assistance, and transitional housing (Burnett et al., 2015; Yu, 2015).

Researchers have attempted to explain whether shelter workers experience stress and the strategies that the shelters implement to help them handle stress. Of particular interest is whether abused women are satisfied with interventions (Lian and Shen, 2014; Wang, 2006), the effectiveness of social workers (Shen, 2009), follow-up services after leaving the shelter (Tutty, 1996), shelter management (Huang, 2015), shelter workers’ initiative to provide support to abused women (Bennett et al., 2004; Bukowski and Buetow, 2011), or some combination of these (Yu, 2008). Although problems regarding shelters have become a prominent subject in social work, the literature on this subject, as well as that on educated opinions among shelter workers concerning these problems, remains limited. Little is known about how shelters are supported, and hardly any information regarding shelters’ difficulties has been gathered. The literature on shelters has taken the perspective of abused women and ignored the role of shelter workers. Thus, many questions remain unanswered. The current study fills the gaps in the literature and makes considerable strides towards addressing pressing challenges identified by shelter workers.

In this study, we sought to gain a more comprehensive understanding of shelter workers’ perspectives by considering 26 shelters for abused women across Taiwan. Investigating the challenges faced by shelters from shelter workers’ (including social workers and supportive workers) perspectives is crucial, because shelter workers have a different position and identity from authorities and abused women. First, shelter workers often tend to be quiet and are often ignored by authorities. Second, the authorities function to help shelters operate in a manner that is consistent with governmental regulations and should not offer limited resources or exert excessive control (Lapierre, 2010; Yu, 2013). Third, the difficulties experienced by shelter workers differ from those experienced by abused women. Thus, shelter workers’ awareness is different from that of their clients.

This article presents an integrative and critical review of studies on shelters with a view to stimulating constructive dialogue regarding challenges faced by shelters. Thus, this article constitutes a crucial first step towards addressing the following long-term goals: (a) meeting shelters’ requirements and (b) improving shelter workers’ performance. The research questions that guided this study are as follows:

What difficulties do shelter workers believe shelters face?

How do these difficulties affect shelters?

Literature review

Taiwan’s women’s shelters, opened in 1992, are official services specifically developed for abused Taiwanese women (Chao, 2005; Kao, 2007; Ko, 2003; Yu, 2008) and remain the primary resource for protecting such women and their children. In many cases, women’s shelters are a last resort for abused women, who often arrive at shelters for want of any other living arrangement.

Although much research on experiences of abused women residing at women’s shelters has been conducted over the preceding few decades (e.g. Beattie and Hutchins, 2015; Lian and Shen, 2014; Peled and Dekel, 2010; Robinson, 2003; Westmarland and Kelly, 2013), little is known about the challenges faced by shelter workers in their demanding and complex roles and what they perceive to be the challenges faced by women’s shelters. Yu (2008) – one of the few researchers to have explored these challenges in Taiwan – noted a feeling of powerlessness among shelter workers, resulting in many such workers leaving the profession and shelters having difficulty employing new staff. This is an inherent problem for shelters and in this profession, which demands that workers operate in a diverse, multifunctional, emotionally intensive, crowded and confidential space; operate with limited governmental resources; receive low salaries and few benefits; work in a stressful environment; experience a relentless workload, burnout, difficulty in setting boundaries and insufficient security; address complex behavioural and emotional problems of mostly short-term residents; deal with a sense of despair at seeing women arrive at the shelter accompanied by children; and manage insufficient time for lobbying and fighting the often conflicting demands of direct practice and ideology (Hombrados-Mendieta and Cosano-Rivas, 2013; Jackson, 1997; Kim et al., 2011; Lian and Shen, 2014; McFadden et al., 2015; Shen, 2009; Yu, 2003, 2008).

More recent studies have supported Yu’s early research, demonstrating that although abuse victims must be cared for, the needs and difficulties of caregivers (shelter workers) and advocacy organisations (shelters) should also be considered (Huang, 2015). Thus, a limitation in our understanding of the opinions and experiences of shelter workers is the inaccessible groups across shelters. Given this challenge, researchers face various vulnerabilities in investigations and distinct barriers to overcoming the working difficulties of shelter workers. Few studies have focused on women’s shelters with respect to the difficulties faced by shelter workers. Even studies that have investigated the difficulties faced by women’s shelters (e.g. Beattie and Hutchins, 2015; Lian and Shen, 2014; Robinson, 2003) have used only quantitative statistics to objectively describe women’s shelters. Thus, these studies have mainly focused on statistical data, such as administration rates, types of services and number of beds. Although these studies provide information, they also leave gaps in our knowledge about shelter workers’ opinions of their working challenges. This article addresses this knowledge gap by presenting their opinions and some collective findings.

Shelters for abused women commonly serve several roles concurrently (Chao, 2005; Kao, 2007; Krane and Davies, 2007; Lapierre, 2010). First, they serve as an immediate safe place for women and sometimes their children. Second, they offer psychological and emotional support pertaining to domestic violence. Finally, an expanded view of the role of shelters considers follow-up services to help abused women transition back into the community. These roles have been considered as competing for the usually limited resources available to women’s shelters (Bukowski and Buetow, 2011; Shen, 2009; Smith, 1989; Wang, 2006), leading to shelter workers facing particularly demanding professional tasks. This article aims to expand the understanding of the experiences of workers in women’s shelters by exploring their perceptions of shelter life and difficulties faced by shelters.

Shelter workers must address the effects of the government on a daily basis. When an evaluation is imminent, shelters must prepare documents and presentations; such preparation may take several weeks. In this context, shelters are all too familiar with the potential outcomes of relationships between shelters and local government, and thus are acutely aware of the opportunity they have to make a difference through evaluations (Huang, 2015; Yu et al., 2016). Shelters may consider their mission to be a matter of life and death, and hence they understandably follow all regulations imposed by the governing authorities, even if some are impractical or unsuitable such as caps on workers’ salaries and inappropriately sized shelter buildings (Tarar and Pulla, 2014; Tutty, 1996; Yu, 2014). Shelters have little power and are unskilled at playing the political game. Most actions taken by shelters focus on raising women’s awareness of control tactics employed by abusers, the effects of such tactics and the cyclical nature of abuse. However, under consideration of the situation they face, shelters can be viewed as requiring the same awareness of control tactics employed by the authorities (Beattie and Hutchins, 2015; Robinson, 2003).

Most related studies have been regional and have used data provided by the authorities; researchers should not rely solely on data from such sources, which may be insufficient for studying all women’s shelters in Taiwan (Lian and Shen, 2014). Therefore, in order to improve shelters throughout Taiwan, a regular investigation method should be established for women’s shelters. Also, Yu (2008) stated that the introduction of shelters to fulfil public needs should be accompanied by regular nationwide surveys, including investigation of the working difficulties faced by shelter workers. This will help address their needs, irrespective of whether they are universal.

Research methods

Participants

A total of 19 small focus groups were established that included service providers (such as managers, social workers and supportive workers) working in Taiwan’s women’s shelters and authority members (such as representatives of the Department of Protective Services and local governments). This study used a semistructured method to allow participants to respond flexibly to the questions posed and to guide the interview and share their views and experiences. This approach enabled us to explore a varied, shifting landscape and to capture information that was local and contextual. The focus groups were established to understand shelter workers’ attitudes towards challenges in shelter operations.

All 26 of Taiwan’s women’s shelters (Table 1) were visited between 1 August 2016 and 11 November 2016. Northern Taiwan has nine shelters (i.e. Taipei City, New Taipei City and Taoyuan City with two each and Hsinchu City, Keelung City and Miaoli County with one each); central Taiwan has five shelters (i.e. Yunlin County with two and Taichung City, Nantou County and Changhua County with one each); southern Taiwan has eight shelters (i.e. Chiayi City and Chiayi County with one each and Tainan City, Kaohsiung City and Pingtung County with two each); eastern Taiwan has three shelters (i.e. Yilan County, Hualien County and Taitung County with one each); and Taiwan’s outlying islands have one shelter (i.e. Penghu County). Each focus group shelter visit lasted approximately 2–3 hours. We had intended to establish 26 focus groups, but because of our limited time and budget, we combined focus groups whenever two shelters were present in one city or county. A total of 21 Health Promotion Administration (HPA) representatives, 50 local government representatives and 94 shelter workers were invited to participate in this study. Finally, 6 HPA representatives, 43 local government representatives and 84 shelter workers were part of the focus group.

Table 1.

Location of Taiwan’s women’s shelters.

Location	Number of shelter	Number of beds
North	9	173
Central	5	59
South	8	87
East	3	34
Outlaying island	1	12
Total	26	365

Focus groups

The focus group schedules (Appendix 1) for the shelter workers were designed to provide participants with the opportunity to discuss the barriers they face in their work in a shelter, what they require to provide services that are more suitable to victims, and what support they receive from the authorities.

Data analysis

The 19 small focus group interviews were digitally recorded. Qualitative data from all the focus groups were transcribed and then thematically analysed (Patton, 2017). Representative selections of participants’ quotes were included within the report to illustrate themes. All data were anonymous and, where necessary, identifying information was removed.

Reliability of the coding themes was established by interrater reliability. The second rater was a graduate student in social work with previous experience in social work research. Using the coding scheme for the three themes established by the student researcher, the second rater coded a random sample of 20 percent of the shelter workers’ responses to the focus group question. On the basis of this sample, the coding of statements by the second rater and the primary researchers was in 82 percent agreement, which is considered acceptable.

Ethics

Ethical approval for this study was granted by Taiwan’s Ministry of Health and Welfare HPA representatives. To convey the purpose of the project and study procedure to the shelters, the researchers held a research orientation session in Taipei for all 26 shelters. In the orientation, the researchers adequately explained and guaranteed that the project was not a performance review of the shelters. Rather, the purpose of the study was to provide information to central and local governments about the challenges faced by the shelters and to gather opinions from shelter workers. Furthermore, we explained that representatives from the HPA would participate in the focus groups. Ultimately, 26 shelters agreed to participate in the study.

To ensure data subject confidentiality, all data were anonymised before use in the research. Pseudonyms were assigned to all participants to protect their identity. To further protect identity, the shelters where participants work were not named. Rather, we used ‘TS’ followed by a number in the results, that is, ‘T’ for Taiwan, ‘S’ for shelter, and a number to describe visit order (e.g. TS01).

Results

In describing the challenges that shelters face, shelter workers made statements that were categorised into one of two themes: (a) statements pertaining to resources and (b) statements pertaining to the Ministry of Interior’s reference standards for regulations governing the shelters. A type of qualitative data display that organises data on the basis of conceptual themes is described as follows (Patton, 2017).

Theme 1: Resources

Participants’ statements concerning resources were categorised into five major areas: (a) lack of staff, (b) insufficient space in the shelter, (c) lack of medium- and long-term shelters, (d) lack of follow-up services and (e) lack of facilities for disabled people. The most frequently mentioned resource concern was the lack of staff. An example of such a statement was, ‘Our current challenge is that we must recruit one more support worker to meet labour law, but no one has applied for the position’ (TS02). ‘It is difficult to recruit support workers because they must work shifts and their compensation and benefits are not really appealing’ (TS05). Many shelter workers revealed further difficulties that stem from this resource challenge, such as the difficulty for social workers to participate in training because of lack of workers. Shelter workers viewed this as damaging, stating that it hindered establishing connections with other professionals and shelters and reduced self-education.

‘We don’t have enough space, so sometimes we’ve got to transfer women to other shelters’ (TS18). ‘Our shelter has no single bedrooms, so sometimes we can’t meet women’s expectation’ (TS13). Participants stated that the shelter had insufficient beds to accommodate women, and concerns about space planning were raised. Currently, all Taiwanese women’s shelters count capacity by the number of beds rather than the number of rooms. Participants suggested that providing single bedrooms was vital because it could ensure abused women and their accompanying children have private space without sharing with others and could boost abused women’s confidence and sense of security. Giving shelters more space and spaces with multiple types of rooms would enable shelters to care for their clients more easily.

Third, participants explained why medium- and long-stay shelters should be established and what could be improved in the meantime. Shelter workers mentioned the importance of ‘establishing medium- and long-term shelters for abused women instead of focusing only on emergency and short-term shelter’ (TS03), and that ‘it is difficult for us to provide in-depth services because women do not stay in our shelter long enough’ (TS26). Participants stated that being unable to ‘provide medium- and long-term shelter services increased the likelihood of abused women continuing their abusive relationships’ (TS20). This increases the loss of efforts the shelter workers had made previously.

Because most shelters are emergency and short-term shelters, no follow-up services are provided to help women achieve self-reliance. Shelter workers stated that ‘some women don’t want to leave the shelter because they are afraid of having no resources and support after leaving’ (TS04) and ‘even though the women want to be self-reliant, we don’t know where to transfer them’ (TS07). Shelter workers suggested that this was essential because the lack of follow-up support services for women means they must return to their abusive relationships. This not only increases difficulty in following the situation of victims, but also increases shelter workers’ working risk.

Finally, shelters worker stated that facilities for disabled people were vital. Numerous participants simply said that shelters should provide barrier-free spaces so that they can accept women with disabilities. Others were more descriptive and suggested necessary modifications that should be made to cater to people with disabilities. Most commonly, participants stated that shelters had no barrier-free toilets, barrier-free ramps, or accessible rooms. Participants described a lack of facilities for the disabled, which caused substantial stress when helping women with disabilities; in some cases, shelters were unable to accommodate them.

Theme 2: Government regulations affecting shelters

These sub-areas included regulations concerning shelters; some shelters did not accept children who accompanied their mother and many shelter workers received low wages and received no night shift differential. In addition, the impact of Taiwan’s new Labour Standards Act was discussed. ‘Government regulations affecting shelters’ seemed an appropriate heading for this coding category because numerous participants stated that government regulations for shelters should be revised, for example, ‘the regulations are inadequate for our actual situation’ (TS25).

Many participants explained the deficiencies in the regulations. For example, ‘it is difficult and risky to accept accompanying teenage boys’ (TS14) and ‘accepting accompanying boys younger than 12 is safe for the shelter’s women’ (TS18). Many participants specifically stated that they suffer from low wages (e.g. support workers’ wages are set in the regulations at approximately NT$25,000, or £625, per month). Living in modern cities, including Taiwan’s capital Taipei, is thus unaffordable. In some cases, this meant ‘a high turnover rate’ (TS11) or ‘difficulty recruiting staff’ (TS19). Finally, participants stated that the new Labour Standards Act had greatly influenced shelters’ management and costs. Numerous participants welcomed the government’s decision to limit employees’ working hours (now set at a maximum of 48 hours per week), and the policy is legally binding and requires that employees are allowed 1 hour of rest after 4 consecutive hours of work. However, ‘we know that the government means well, but their new laws have been no help to us’ (TS16), and ‘the new law means that shelters require more staff’ (TS10). Participants opined that the shelters were unable to ‘abide by the new law because the shelters do not have enough staff to assign work to and reduce overtime’ (TS06), and found ‘recruiting new professionals like social workers and support workers difficult in such a short time and under budget constraints’ (TS20).

Although the content of the Labour Standards Act is beyond the scope of this study, all public- and private-sector entities are bound by it and must adjust their management accordingly. However, many shelter workers suggested that the Act had disastrous consequences. By contrast, several participants advocated revising the act and mentioned no benefits of the Act. In particular, participants noted that shelters are supported by the public and private sectors to provide shelter services. This might be particularly difficult for shelters that cannot actively lobby for their own interests and the interests of their residents. Clearly, this topic should be examined more thoroughly. The data from this study allow for definite conclusions to be made.

Discussion

Shelters for abused women have been successful in offering support and safety for women and enabling them to contemplate a route to living without violence (Krane and Davies, 2000; Peled and Dekel, 2010). However, to date, little specific attention has been directed towards investigating the particular interests and needs of shelters. As this article indicates, the physical and emotional burdens of shelter workers are usually well taken into account. When limitations are noticed, they are either romanticised or scrutinised through the lens of power and control relations, which are central to the understanding of many interventions against shelters. Heightened sensitivity towards the intervention of power coupled with unexamined and idealised assumptions of authorities’ care may produce inaccurate understanding of cooperative relations in the context of responding to shelters’ situations.

As demonstrated in this study, shelter workers continually face multilayered clashes between personal, organisational and sociocultural perceptions of abused women as clients. Therefore, the task of such workers in finding a balanced response to services and job problems at women’s shelters is a demanding and complex one and can cause workers to experience stress and a sense of powerlessness. This may partially account for negative outcomes of overwork with victims of domestic violence due to a lack of staff sharing the workload (Huang, 2015; Yu, 2015). Studies on this topic have discovered high rates of working fatigue and burnout, as well as a wide range of posttraumatic stress, depression, anxiety and general psychological distress symptoms, as well as changes in perceptions of the self and of others (e.g. Hombrados-Mendieta and Cosano-Rivas, 2013; Kim et al., 2011; McFadden et al., 2015).

In terms of experiences with abused women and their children in shelter settings, shelter workers are typically aware of insufficient space. Conceptions of helping abused women transition from preventing abuse as a necessity in terms of having medium- and long-term shelters are significant. Facilities for disabled women and follow-up services must be installed and consciously distinguished from other facilities that are part and parcel of shelter service (Kao, 2007; Ko, 2003; Smith, 1989; Yu et al., 2016). Shelters in this field may find it difficult to offer such arrangements. Not only must shelters challenge the authorities that provide them with resources, they should also widen their range of operation, for example by inviting authorities and social work scholars to visit them at fixed or nonfixed times.

In an effort to follow organisational requirements, shelters may emphasise women’s responsibility to function adequately as clients, namely to control their children and provide for their children’s emotional and physical needs. The ideology and practice guidelines of most shelters, including those in Taiwan (Ministry of the Interior, 2017), direct shelter workers to persistently and continually improve the empowerment of residents in an effort to assist them in regaining a sense of security, confidence and control over their lives (Krane and Davies, 2007; Lapierre, 2010; Peled and Dekel, 2010; Westmarland and Kelly, 2013). Dilemmas arising from the need to comply with these often contradicting organisational practices as they relate to women’s autonomy were raised repeatedly in the group interviews. Shelter workers mentioned the paradox inherent in empowering women when no private family rooms or spaces are available. Women’s shelters may be unable to offer sufficient space for abused women and their children. Shelter workers may need to more cautiously help women anticipate the challenges that await them in their transition back into society to further their sense of autonomy.

Our results exhibited similarities with results from studies in which the barrier and difficulty profiles of shelters have been constructed. Similar to Yu (2004), we determined that most challenges reported by our participants were related to insufficient resources, reflecting an apparent lack of funding from public and private sectors. Participants reported that the numbers of professional staff were being reduced. Therefore, the effectiveness and quality of shelter services could not be guaranteed (Huang, 2015; Kao, 2007; Yu, 2008). The role of professional social workers and supportive staff in shelters must first be reviewed; we believe that such a review may open up possibilities for greater engagement with research on professionalisation with respect to social work. Therefore, further research could investigate the extent to which notions of effectiveness and quality shape practices in a situation with too few professional shelter workers. Resources were a significant predictor in the present study. Perhaps, its significance was confounded in our results because most shelters temporarily increased shelter, or it may suggest evidence of a location difference. Funding, which was vital in every prior study, although not always in the same situation, was also vital in this study, suggesting that funding can reflect different governments’ attitude towards support for shelters.

Our results also point to the fact that women’s shelters must lobby for changes to regulations governing shelters, such as increasing workers’ low wages and the duty allowance of shelter workers. Reasonable wages and meeting the requirements of Taiwan’s Labour Standards Act would increase willingness to work at a shelter, and would probably be most beneficial for the most disadvantaged shelter workers because they have expressed the strongest need for help with central and local government funding. In agreement with previous studies (e.g. Arnold and Ake, 2013; Burnett et al., 2015; Tarar and Pulla, 2014), our data suggested that shelters may be expected – or shelters themselves may wish – to promote gender equality and women’s rights and provide child protective services and follow-up services through a range of efforts and social advocacy. However, such tasks conflict with the tight funding available to shelters (Bukowski and Buetow, 2011; Shen, 2009; Smith, 1989; Wang, 2006). Furthermore, although the current regulations allow entry of anyone under 18 years into a shelter, regardless of gender, some shelters permit only children under 12 years, regardless of gender, to prevent insufficient rooms or beds because of a limited budget (Yu, 2013). Therefore, balancing adherence to the law and acceptance of abused women’s children is critical to women’s responses to the services offered by shelters, encouraging shelter workers to simplify their tasks and operate effectively (Yu, 2015).

Implications for practice and policy

Several implications for practice may be drawn from this study. Shelter workers may benefit from new staff being recruited, space being added to the shelter and methods of balancing or resolving existing difficulties being found. Such changes may render shelter workers more capable of balancing all factors in response to work-related problems and avoiding work-related stress. For example, staff may require a clear definition of the boundaries they are expected to adhere to when looking after their clients at the shelter in light of the tension between workload and feelings of powerlessness (Beattie and Hutchins, 2015; Hombrados-Mendieta and Cosano-Rivas, 2013), and the development of identification and countertransference with the authorities (Chao, 2005). Problems related to inadequate resources – such as insufficient space, a lack of follow-up services for abused women and a lack of facilities for disabled people at shelters – can be attended to in organised forums such as symposia, authority meetings and external introspective group gatherings (Wang, 2006); at regular local government supervision meetings, during which individual aspects of the complex relationship between shelters and the authorities can be further processed (Arnold and Ake, 2013; Huang, 2015); and through exposure to the literature regarding the difficulties faced by shelters when helping victims of domestic violence. Staff members can use several methods of self-care to reduce work-related stress, including self-nurturing and support seeking, participating in personal therapy, diversifying their professional responsibilities, and participating in political action as a means of empowerment and fighting for more support for shelters (Burnett et al., 2015; Yu, 2004).

The central and local governments responsible for funding shelters are diverse and include social services, health and children’s support departments. In many locations and at the central level, other ministries offer services relevant to domestic abuse. Each such ministry has a mandate much broader than solely providing safety for abused women; most are also required to promote public awareness and fund prevention programmes. From the governmental perspective, while the authorities must be committed to protecting abused women at shelters, they are also responsible for shelter funding.

A major concern regarding policy is that the regulations governing shelters should be adjusted. Low wages are one problem; wages have been frozen for a number of years or low wages have become a normality. Because funding has not increased, special programmes have been cut despite shelters emphasising their importance. Therefore, the wages and duty allowances of shelter workers should be increased through regulation, and shelters should recruit new staff rather than relying on only current staff to mitigate the effects of the new Labour Standards Act. This might mean stressing to shelters the importance of lobbying the central and local governments. Lobbying the authorities to inform them about the deficiencies in the regulations governing shelter workers could be a long-term endeavour. This suggests that shelters should explain any barrier that affects their operations (Rasool, 2016; Yu, 2008).

In addition, large changes in policy execution affect what services must be offered. For example, some shelters hesitate to accept accompanying boys older than 13 years. This violates the law, which states that shelter must accept girls and boys under 18. Therefore, discriminating against children on the basis of age and sex is believed to reduce victims’ willingness to stay in a shelter (Lian and Shen, 2014; Yu, 2003). With expanded awareness of women’s rights largely due to the efforts of shelter advocates, new services and legislation that will likely enable shelters to help women and children feel safe and supported have been developed. This partially acknowledges that most abused women go to shelters accompanied by their children (Bukowski and Buetow, 2011; Lian and Shen, 2014; Shen, 2009; Smith, 1989; Wang, 2006). Shelters must follow the regulations related to this matter because women might refuse to seek shelter in transition houses out of fear that their children may be taken and from their care.

Conclusion and limitations

Promoting women’s shelters, understanding the needs of abused women, and helping women shelter from violent situations are the work of a lifetime. The findings of this study serve as a foundation for improving and developing interventions for women’s shelters and their staff. Improving Taiwan’s response to the challenges faced by women’s shelters requires fundamental changes in societal attitudes, professional practices and authoritative systems. Until helping strategies are coordinated and empirical findings are consistently applied, shelters and their staff will continue to suffer. Social workers and scholars should remember the words of one young shelter worker: ‘We often feel powerless even though we would like to do more for our clients … under the authorities’ regulation and control … we really do not know what to say or do.’ Our study allowed shelter workers to voice their perspectives and provided the public and the authorities with an understanding of the concerns affecting shelters and their workers. Shelters deserve more resources, and their workers deserve less stress at work.

Shelter workers could lobby to secure essential rights such as lower workloads, achieved by further hiring of staff and enabling of swift response to clients’ needs. If such changes are applied, shelter workers will not need to frequently work overnight shifts and will have more leisure time to relieve stress. In addition, workers could have more time with their families and improve their professional performance by attending social worker training. Relevant supervision of women’s shelters and appropriate loosening of regulations, including those regarding appropriate working conditions (i.e. salary increases), may help to create an environment where the aforementioned goals can be achieved. From another perspective, the findings regarding limited space for abused women and facilities for disabled women at women’s shelters indicate that local governments must be made aware of the importance of available space at such facilities. Equally crucial are follow-up services such as safe plans in the community, workshops and client visits to prevent abused women from being re-exposed to domestic violence. Moreover, emphasis should be placed on improving the working conditions of staff, meeting the needs of shelters, and increasing the autonomy of shelters, which has been disregarded in recent years.

The strengths of our study were the inclusion of numerous study samples and different locations. The current study provided practical results regarding the aspects of staffing that affect shelter situations; these results indicate the importance of continuing to improve services for abused women and the operation of shelters. More detailed understanding should be attained regarding not only how workers in shelters for abused women differ from other types of shelter workers, but also how shelters differ from one another. Why some types of shelters have a high staff retention rate whereas others are consistently understaffed is unclear. Answering such questions necessitates continued research on the characteristics of shelter operations, and the methodological concerns presented in this article remain of great importance. This study did not obtain many answers, but rather sought to raise questions and initiate a more in-depth discussion about social work. Potential contributions are shedding light on the limitations of previous studies and continuing to strive to find methods of helping abused women at shelters.

Regarding interpretation of our results, limitations of this study should be considered, including our generalisation and the use of focus groups to collect data. Although the research sample was heterogeneous, it may not have been representative of the entire population of shelter workers in Taiwan or elsewhere. This, combined with the nature of the qualitative methodology, implies that the findings should be transferred to other contexts with caution (Patton, 2017). Although the group format encouraged the participants to discuss and respond to shared experiences and roles, the shelter workers’ perspectives and responses regarding shelter practices may have been influenced by the emphasis placed – both during shelter visits prior to the focus group interviews and during the interviews themselves – on perspectives of difficulties with serving clients at the shelter. The shelter workers’ critical views of the status of shelters may have been reinforced by their extensive reflection on the state of women who have suffered from domestic violence and those who reside in shelters.

Footnotes

Appendix 1

Focus group schedule – shelters’ workers

Acknowledgements

We thank all of the shelter workers and authorities who participated in this study.

Declaration of conflicting interests

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

Funding

We thank the Department of Protective Services, Ministry of Health and Welfare for their funding and constructive and valuable support.

ORCID iDs

Min-Yu Liao

Mei-Kuei Yu

Author biographies

Min-Yu Liao is an assistant professor at the Chinese Culture University, Taipei. She has research and teaching interests in violence against women, child protection, school social work and social group work. She currently engages in several women’s welfare organisations and family service centres as an external supervisor.

References

Arnold

Ake

(2013) ‘Reframing the Narrative of the Battered Women’s Movement’, Violence Against Women 19(5): 557–78.

Beattie

Hutchins

(2015) ‘Shelters for Abused Women in Canada, 2014’. Available online at: http://www.statcan.gc.ca/pub/85-002-x/2015001/article/14207-eng.htm (accessed 13 April 2017).

Bennett

Riger

Schewe

Howard

Wasco

(2004) ‘Effectiveness of Hotline, Advocacy, Counseling, and Shelter Services for Victims of Domestic Violence: A Statewide Evaluation’, Journal of Interpersonal Violence 19(7): 815–29.

Bukowski

Buetow

(2011) ‘Making the Invisible Visible: A Photovoice Exploration of Homeless Women’s Health and Lives in Central Auckland’, Social Science & Medicine 72(5): 739–46.

Burnett

Ford-Gilboe

Berman

Ward-Griffin

Wathen

(2015) ‘A Critical Discourse Analysis of Provincial Policies Impacting Shelter Service Delivery to Women Exposed to Violence’, Policy, Politics, & Nursing Practice 16(1–2): 5–16.

Bybee

Sullivan

C.M.

(2005) ‘Predicting Re-Victimization of Battered Women 3 Years After Exiting a Shelter Program’, American Journal of Community Psychology 36: 85–96.

Chao

(2005) ‘A Study in Social Change: The Domestic Violence Prevention Movement in Taiwan’, Critical Asian Studies 37(1): 29–50.

Hombrados-Mendieta

Cosano-Rivas

(2013) ‘Burnout, Workplace Support, Job Satisfaction and Life Satisfaction among Social Workers in Spain: A Structural Equation Model’, International Social Work 56(2): 228–46.

Huang

S.L.

(2015) ‘Reflecting on the Practices of Taiwan’s Family Violence Victim’s Shelters through some Lessons Learned from a Center for Family Violence Victims in Hong Kong’, Asian Journal of Domestic Violence and Sexual Offense 11(1): 137–54.

10.

Jackson

A.K.

(1997) ‘Domestic Violence in Northamptonshire: Fifty-Five Women’s Experiences’, Unpublished Master’s Thesis, University of Leicester, Leicester, UK.

11.

Jarvis

K.L.

Gordon

E.E.

Novaco

R.W.

(2005) ‘Psychological Distress of Children and Mothers in Domestic Violence Emergency Shelters’, Journal of Family Violence 20(6): 389–402.

12.

Kao

F.H.

(2007) The Domestic Violence Prevention and Treatment Act. Taipei: Wu-Nan Culture.

13.

Kim

Kao

(2011) ‘Burnout and Physical Health among Social Workers: A Three-Year Longitudinal Study’, Social Work 56(3): 258–68.

14.

S.L.

(2003) ‘Roles Shelters Play throughout Battered Women’s Recovery Process – Reflection on Taiwan’s Experience’, Community Development Journal 101: 310–9.

15.

Krane

Davies

(2000) ‘Mothering and child protection practice: Rethinking risk assessment’, Child & Family Social Work 5(1): 35–45.

16.

Krane

Davies

(2007) ‘Mothering Under Difficult Circumstances: Challenges to Working with Battered Women’, Journal of Women and Social Work 22(1): 23–38.

17.

Lapierre

(2010) ‘More Responsibilities, Less Control: Understanding the Challenges and Difficulties Involved in Mothering in the Context of Domestic Violence’, The British Journal of Social Work 40(5): 1434–51.

18.

Lian

T.T.

Shen

C.T.

(2014) ‘An Outcome Evaluation on Protective Service for Victims of Intimate Partner Violence’, Asian Journal of Domestic Violence and Sexual Offense 10(2): 1–34.

19.

McFadden

Campbell

Taylor

(2015) ‘Resilience and Burnout in Child Protection Social Work: Individual and Organisational Themes from a Systematic Literature Review’, The British Journal of Social Work 45(5): 1546–63.

20.

Ministry of Health and Welfare (2017) ‘Statistics of Women Welfare Services’. Available online at: https://dep.mohw.gov.tw/DOS/cp-2975-14657-113.html (accessed 7 April 2017).

21.

Ministry of the Interior (2017) ‘Government Information Disclosure’. Available online at: http://www.moi.gov.tw/chi/chi_public_info/index.aspx (accessed 7 April 2017).

22.

Patton

M.Q.

(2017) Qualitative Research & Evaluation Methods. Thousand Oaks, CA: SAGE.

23.

Peled

Dekel

(2010) ‘Excusable Deficiency: Staff Perceptions of Mothering at Shelters for Abused Women’, Violence Against Women 16(11): 1224–41.

24.

Rasool

(2016) ‘Help-Seeking after Domestic Violence: The Critical Role of Children’, Journal of Interpersonal Violence 31(9): 1661–86.

25.

Robinson

A.L.

(2003) ‘The Cardiff Women’s Safety Unit: A Multi-Agency Approach to Domestic Violence’. Available online at: http://www.academia.edu/11144772/The_Cardiff_Womens_Safety_Unit_A_Multi-Agency_Approach_to_Domestic_Violence (accessed 12 April 2017).

26.

Shen

C.H.

(2009) ‘The Study of Practical Barriers of Social Workers for Marital Violence’, Social Policy and Social Work 13(2): 87–142.

27.

Smith

L.J.F.

(1989) Domestic Violence: An Overview of the Literature. London: Home Office.

28.

Tarar

M.G.

Pulla

(2014) ‘Patriarchy, Gender Violence and Poverty amongst Pakistani Women: A Social Work Inquiry’, International Journal of Social Work and Human Services Practice 2: 56–63.

29.

Tutty

L.M.

(1996) ‘Post-Shelter Services: The Efficacy of Follow-Up Programs for Abused Women’, Research on Social Work Practice 6(4): 425–41.

30.

Wang

S.Y.

(2006) ‘Working in a Battered Woman Shelter in Taiwan: Stress and Coping Strategies’, Social Policy and Social Work 10(1): 189–226.

31.

Westmarland

Kelly

(2013) ‘Why Extending Measurements of “Success” in Domestic Violence Perpetrator Programmes Matters for Social Work’, The British Journal of Social Work 43(6): 1092–110.

32.

M.K.

(2003) ‘A Comparative Study between Taiwan and England on Women’s Experiences of Domestic Violence and of Services Delivery Systems’, Unpublished Doctoral Thesis, University of Kent, Canterbury, UK.

33.

M.K

. (2004) ‘Abused Women’s Experiences of Refuge Services in Taiwan and England’, Taiwanese Journal of Social Welfare 3(2): 117–158.

34.

M.K.

(2008) ‘Research on the Transaction of Refuge Services for the Abused Women in Taiwan’, NTU Social Work Review 18: 143–90.

35.

M.K.

(2013) A Shelter’s Guide to Theory and Practice for Battered Women. Taipei: Hungyeh Publishing.

36.

M.K.

(2014) ‘Elucidating Service Program Responses to Domestic Violence in Taiwan’, NTU Social Work Review 29: 53–96.

37.

M.K.

(2015) Preventing Family Violence: A Social Worker’s Guide to Theory and Practice for Survivors. Taipei: Hungyeh Publishing.

38.

M.K.

Cheng

L.C.

Chang

H.Y.

, et al. (2016) ‘A Promoting One-Stop Service for Domestic Violence’, Community Development Journal 156: 293–301.