Domestic violence (DV) shelters (women’s refuges) play a crucial role in responding to the needs of women and children subjected to DV. They are often seen as safe places offering protection, and a stay at a DV shelter can improve women’s perceived safety, mental health, and stress, and generally has few disadvantages (Yakubovich et al., 2022). Children often accompany their mother to a DV shelter, and they have generally experienced substantial violence too. Among children living in DV shelters, 80–90% have witnessed violence and about 50% have themselves been the direct targets of violence at home (Fernández-González et al., 2018). Such violence increases the risk of problems for the child, ranging from emotional issues, such as anxiety, depression, and social withdrawal, to behavioral challenges, such as violence and other norm-violating behaviors (e.g., Sternberg et al., 2006; Vu et al., 2016). These risks highlight the need to protect children from DV, and shelters can provide such protection. Against this backdrop, our study focuses on children’s experiences of staying at DV shelters.

Most children appreciate the safety that a DV shelter can offer (Chanmugam, 2011; Øverlien, 2011a; Thunberg et al., 2022). Research has shown that, in recent years, DV shelter workers have adopted a child-centered perspective that considers the children’s situations during their stay and supports them alongside their mothers (Côté et al., 2022; Eriksson, 2010). Children themselves state that the staff are nice (Øverlien, 2011b). In a study of young adolescents residing in DV shelters, Chanmugam (2011) also found that the relational environment and shelter milieu itself, including the other families residing there and the sense of camaraderie, were important to them.

However, staying at a DV shelter also brings challenges for children. For example, they often experience shelter life as passive and slow with rigid rules, and they are frequently unable to attend school (Chanmugam, 2011; Øverlien, 2012). Children also often express a dislike of shelter rules, which restrict them, including rules pertaining to curfew, bedtime, and parental supervision (Chanmugam, 2011). Adding to this, Thunberg et al. (2022) found in their review that restrictions regarding safety and security can adversely affect other areas of these children’s lives, such as the opportunity to go to school or to engage in after-school activities, and social relationships. Additionally, Pinto et al. (2019) found that children living in shelters displayed higher levels of both internalizing and externalizing symptoms compared to children living at home with their mothers and abusive partners. Thus, it cannot be taken for granted that DV shelters are solely positive for children.

Moreover, researchers like Øverlien (2011b) and Vass and Haj-Yahia (2021) have shown that children who move to DV shelters with their mothers lack information and hence struggle to understand why they had to leave their home, where they were going, or what they should take with them. They also worry about their mother’s emotional state and the emotional reactions of siblings, as well as what might happen when they leave the shelter. All of this may render them confused and anxious. After moving to a DV shelter, children may also struggle to maintain bonds with their parents. After the move, they may have limited opportunities to experience the bond with their father, while the physical context of the shelter enables children and mothers to engage in emotional and embodied practices, such as hugs and shared activities (Henze-Pedersen, 2021). However, Vass and Haj-Yahia (2023) found that previous exposure to violence against their mother continues to affect children, even during their stay in the shelter. They are affected not only by the violence itself but also by the consequences of their mother’s experiences, which can negatively influence children’s well-being and their relationship with their mother (Vass & Haj-Yahia, 2023).

In addition, some children face multiple stays at DV shelters. This may be due to prolonged exposure to DV, and may lead to disruptions to close relationships and repeated disruptions to preschool and school attendance (Selvik & Øverlien, 2015). Thus, from an international perspective, the experience of moving to and living in a DV shelter may entail both benefits and challenges for children. It must be remembered that, while protecting children against violence is important, their quality of life depends upon multiple dimensions, which the children themselves highlight. Still, knowledge about children’s experiences of the shelter environment and their quality of life while living at a DV shelter is limited, and prior to our project no research had been conducted on children’s own experiences of staying at a DV shelter in Sweden.

The Swedish Context

In Sweden, approximately one in ten children experience some form of physical violence between their parents at some point while growing up, and many of these children have also experienced some form of direct abuse by an adult (Cater et al., 2015; Jernbro et al., 2023). Both exposure to and the severity of DV increase the risk of symptoms of PTS, anxiety, and depression, as well as self-harming and other non-clinical problems (Cater et al., 2015). Although most violence between parents in Sweden seems to be bidirectional, violence (and verbal attacks) perpetrated by fathers against mothers tends to be more severe and related to poorer long-term mental health for their children (Miller et al., 2014). Therefore, the DV we discuss in this article primarily concerns fathers’ violence against mothers. Children exposed to DV in Sweden typically struggle to make sense of their father’s violence, and their attitudes and behaviors afterwards, because they assume that fathers should be good and caring and that violence is bad, and something one should regret (Arnell & Källström, 2023).

There are no definitive statistics for how many children experience staying at DV shelters in Sweden but, according to the National Board of Health and Welfare (2020), approximately 6,500 adults and 6,200 children stayed at shelters in 2020. The average stay is about two months (National Board of Health and Welfare, 2020), but it can vary from a few days up to as long as a few years. In 2020, 54% of Swedish DV shelters were run by non-profit organizations, 37% as private companies, and 9% by municipalities (National Board of Health and Welfare, 2020). The types of accommodation also vary, and include apartments or shared accommodation, where families generally have their own bedroom but share a kitchen and living room, and there is no demand or recommendation on specific programs for children. This means that the living conditions and experiences of a shelter stay may differ among children in terms of aspects such as social relations at the shelter. In Sweden, DV shelter staff emphasize the importance of creating a secure, dignified, and home-like environment (Ekström & Hvenmark, 2025).

Children’s rights regarding violence and staying at DV shelters in Sweden have been strengthened in many ways during recent years. For example, the rights of children exposed to DV as victims of crime were strengthened in the Social Services Act in 2007 (Chap. 5 § 11, 2001:453), exposing a child to violence against their parent was criminalized in 2021 (Swedish Penal Code, 1962:700), and new legislation to specifically strengthen the rights of children in DV shelters was implemented in 2024 (Act, 2024:79) on Placement of Children in Sheltered Housing). However, how this context and the recent efforts to strengthen the rights of children exposed to DV affect children’s experiences in DV shelters remains largely unknown. Therefore, in this study we aimed to contribute to the field by examining how children describe staying at DV shelters in Sweden, and identifying important aspects, both positive and negative, of such experiences for their quality of life. We achieved this by analyzing interviews with 12 children, aged 7–17 years, using thematic analysis, as developed by Braun and Clarke (2006).

Method

This study is part of the project Article 19: What sheltered housing means for abused children. It is based on 12 interviews with children who have experiences of living at a DV shelter together with their mother, and in some cases siblings, after leaving a violent home. The children, four boys and eight girls, were aged 7–17 when the interviews were conducted, with a median age of 10.6 years. The length of their stay varied between three months and three years. Six of the children had stayed in just one shelter, one child in two shelters, and the other five had stayed in between three and eight shelters. All the children had stayed at shelters with shared accommodation, such as shared kitchen and living room space, but some had also stayed at shelters where the family had their own apartment, and did not share space with other adults or children. Based on the children’s narratives, the shelters with shared accommodation usually had designated play areas, structured activities, and staff with a child focus, while placement in an apartment was less structured and often involved fewer contacts with shelter staff. Eight of the children were still living in a shelter at the time of the interview, and four had recently moved out. The children came from eight different families, and the interviews were conducted individually, focusing on each child’s narrative and their specific experience of living in a DV shelter. Ten children had fled violence perpetrated by their fathers, while two had escaped abuse from their stepfathers.

Recruitment and Consent

Recruitment of children began in late 2020 but was obstructed by the Covid-19 pandemic, which limited access to DV shelters during that period. As a result, interviews were conducted between March 2022 and June 2023, with most taking place after autumn 2022, when the final restrictions in Sweden were lifted. We recruited children with the help of staff at DV shelters in Sweden. The staff informed the mothers about the study, and if the mothers deemed participation suitable, they then informed their children and asked if they were interested. If they were interested in participating, the mothers contacted us, the researchers, directly or via the shelter staff. We mutually agreed upon a time and place to meet, where the mothers and their children felt safe. Then and there, we provided the mothers and children (adjusted to their age) with detailed information about the study in both verbal and written form and answered any questions about participation. If the mother then consented, she signed a written consent form for her child/ren to participate. Then, each child was asked if they wanted to participate, and informed that they could take a break at any time, skip questions, or discontinue the interview if they wanted to. If the child verbally assented to participation, the interview started.

Procedure

We conducted the interviews in a place chosen by the mother, a setting where both she and the child felt safe. They were structured around the five phases that women and children who live in DV shelters go through, as formulated by Rösare (2015): (1) the arrival and first day at the shelter, (2) the first week at the shelter, (3) continuing time at the shelter, (4) the final time at the shelter, and (5) the time after leaving the shelter. However, we employed these themes flexibly in the interviews, allowing the children to talk relatively freely, rather than following a chronological order. To allow the children to tell their own stories, we based the interviews on open-ended questions inviting them to describe their experiences; for example: “Can you tell me what staying at the shelter was like?” or “Can you describe what it looked like at the shelter?” The children’s narratives were then filled out with follow-up questions to gain a deeper understanding of their experiences, and sometimes to validate that we understood their story. During the interviews, we also invited the children to draw pictures if they wanted to and we provided coffee/lemonade and cookies. The interviews lasted approximately 25 to 60 min, depending on how much the child wanted to talk. Typically, if the child was a teenager, the interview lasted longer.

Data Analysis

The interviews were analyzed using Braun and Clarke’s (2006; see also Braun & Clarke, 2013, 2022) six phases of thematic analysis as a framework and guide. The six steps consist of: (1) familiarizing yourself with the data, (2) generating initial codes, (3) searching for themes, (4) reviewing themes, (5) defining and naming themes, and (6) producing the report. These phases can be understood as a process, moving back and forth between the steps, which is also how we conducted this analysis. More specifically, we began the analysis process while the first and third authors were alternating between conducting the interviews and transcribing them (i.e. the person who did not conduct the interview transcribed it) in order to become more familiar with the data. Then, all three of us further familiarized ourselves with the data by reading and re-reading the transcripts to identify initial codes and discussing possible themes that would represent the data well. Because the aim of this study was to explore the children’s experiences at the shelter, we disregarded the parts of the interviews concerning fleeing from the violence and the period after moving from the shelter to their own accommodation. After this, the first author arranged the codes into a shared coding scheme, with reference to each child. Then all three authors reviewed and discussed the themes and their correspondence with the codes until we agreed that they corresponded with the contents of the narratives.

We revised and decided upon the themes, sub-themes, and their names in joint discussions. Finally, in the sixth step, the first author drafted the report of the findings, presented below, describing the children’s narratives of their experiences of DV shelters. We have used excerpts to illustrate the results and we translated these from Swedish into English ourselves, after which they were proofread by a native English-speaking proofreader to whom the original extracts were also available for comparison, with the goal of making the translation as accurate as possible.

Ethical Considerations

We obtained ethical approval from the Swedish Ethical Review Authority (Dnr. 2020:04561, with changes Dnr. 2021-03242, 2021-06928-02 ). Because the project includes children with experiences of violence within the family, and many of them were living with protected identities, we have had to address many ethical considerations. The sensitivity of the project influenced the entire process, from the initial planning, through the recruitment process, data-collection, handling the data, and writing this manuscript. All three authors have previously been engaged in victim support, one as a crime victim support volunteer, one as DV shelter staff, and one as a member of another DV shelter board, but none of us have stayed at a DV shelter as either a child or adult. While conducting this study, we have benefitted from our previous work with victim support and our previous research on young crime victims. Trying to ensure that the findings remained grounded in the children’s lived experiences, rather than our own preconceptions, we have made a conscious effort to focus on participants’ descriptions and meanings, both during the interviews and while conducting the analysis.

During the interviews, for the well-being of the participants, we paid particular attention to the children’s verbal and non-verbal expressions. The mothers and children received written information about where they could turn for help and support. The children were also given the option to have a “safe” person with them during the interview, and three children chose to have their mother present. In addition, the staff at the shelters who had supported us during the recruitment process helped us to meet the mothers’ and children’s need for support and safety. To protect the identities of both mothers and children, we have used pseudonyms in the reporting of our findings below. We have also removed specific details, such as names of places and organizations, to protect the participants’ identities.

Findings

During the coding process, three main themes emerged from the children’s descriptions of staying at a DV shelter: Shelter environment, Activities, and Social relations. The children described life at a DV shelter in both positive and negative terms. They said that the shelter environment, activities, and social relations were important for their quality of life and, when these were lacking, they risked feeling bored and isolated. Those who had lived at more than one shelter tended to describe each shelter quite differently, in terms of both the shelter environment, and the prerequisites for and number of activities and social relations. We present the three themes individually below, with a focus on illuminating both the positive and negative aspects of DV shelter stays as described by the children.

Shelter Environment: Safety, Basic prerequisites, and Isolation

The interviews show that the children highlighted safety and basic prerequisites, such as a clean and hygienic environment, as important in their narratives, but they also talked about negative aspects that risked creating a feeling of isolation. Regarding safety, all the children said that they experienced and appreciated the safety at the shelter. They talked about safety in relation to issues such as security, worrying less, and living in a place that was free from the violence they had experienced at home. As one example, Mattias, a nine year old boy, express that “I felt fine then [before moving to the shelter] but I feel much better here […] It’s much safer and better, and you get to be with the staff.” Another example, showing how the children talked about their wellbeing in relation to safety is from Sebastian, a 17 year old boy, saying “It just feels good to get to an environment like I, I got my own room, so it was more like I got to be by myself, away from everything, so it was just nice.”

These examples from Mattias’ and Sebastian’s narratives illustrate positive aspects of being in a safe place, which were commonly shared by all the children, no matter what age they were, whether the perpetrator was their father or stepfather, or how many shelters they had lived in. Although safety can be understood as a basic prerequisite for children at a shelter, in some cases the children also spoke specifically about the basic prerequisites in relation to the shelters’ standards and cleanliness. When these were lacking, the children described the low standards and an unclean or unhygienic environment as affecting their stay in negative ways, as Lisa’s narrative shows.

It was, well, a small room with maybe, it was three beds, and the shower wasn’t clean at all, so mom had to nag for days that we didn’t want to stay in the room and that we could change rooms because it was so, well, dirty that we didn’t want to shower. I couldn’t shower because we were used to everything being clean and stuff, so we didn’t shower for several days. (Lisa, girl, age 15)

As illustrated by Lisa’s comments, the children’s narratives about a clean and hygienic environment can be understood as describing basic prerequisites for living in a shelter. In addition, the children highlighted the standard of their rooms, the furniture, toys, and play areas as basic prerequisites that were important for life at the shelter. When described in positive terms, such as clean spaces, having a room big enough for the whole family, and a place where they could play, the children talked about the shelters in positive terms, describing them as fun, and saying that they enjoyed staying there. When these basic prerequisites for children’s lives were missing, the experience of life at the shelter was instead often described as having a negative effect, and being dull, boring, and socially isolating, as shown in Freja’s narrative.

I wasn’t allowed to go outside, so that was very sad because it was very, I couldn’t kind of move. It’s been kind of the hardest part because I think it’s fun to be outside. […] And then, now, where I was last, I wasn’t allowed to go to school for six months. (Freja, girl, age 11)

As Freja’s narrative shows, life at a shelter can be experienced as boring and isolating; for example, not being able to spend time outside or go to school. This isolation needs to be understood in relation to the rules and regulations regarding safety that must be followed due to the threat posed to the families. Children’s safety and security, which they value highly, thus need to be weighed against the negative consequences that isolation entails.

To summarize, the children’s narratives reveal the importance of safety and basic prerequisites at the shelters to make their stay a positive experience. Still, their narratives show that living in a shelter can also be experienced as boring and isolating and may have negative effects on their quality of life, especially if there is no opportunity to attend school and a lack of activities.

Activities: Leisure time and Outdoor Activities

The children’s narratives reveal not only the significance of the shelter environment, but also the importance of activities, more specifically outdoor activities and things to play with or do during their leisure time, such as toys, stuffed animals, or computer games, as Mehmet (boy, age 10) highlighted by saying “It’s fun that you can borrow games here, Nintendo and PS4.” However, when these were missing, children described their experiences of the shelter stay as boring, revealing negative aspects of such stays, as exemplified by Elin’s narrative:

I wish they would maybe give us some toys or something so we can at least do something. Or at least help us do some activities or something like that. It doesn’t have to be anything huge, but at least so we can do something and not just have to sit there in the shelter and have a boring time. (Elin, girl, age 11)

The children’s narratives stress the importance of activities and things to do during leisure time, as illustrated by Elin. Another important aspect is the opportunity to go outside, as illustrated by Freja, above, and here contrasted against her experiences at another shelter.

At the last shelter I was in, I was able to go out. There was a courtyard, a huge one, it was like the greatest garden, so huge, which we were allowed to be in, but not outside the border, and then I got a bicycle, so then we cycled a lot and so on. (Freja, girl, age 11)

As Freja’s narrative shows, life at a shelter can be quite fun and enjoyable if it is possible to go outside or engage in activities like cycling, but when forced (for security reasons) to stay indoors it becomes boring and isolated. The available activities, especially the opportunity to be outdoors, are often described in positive words, as Camilla’s narrative shows.

The best thing was that there you could, it was only us who lived there and we got to go on lots of excursions. We went to [name of an amusement park] and we’ve been to the bathhouse, we’ve been to a playground. We did a lot of fun things. (Camilla, girl, age 7)

As these narratives show, activities are of great importance for the children. More specifically, younger children often emphasized the importance of having toys, a trampoline, and designated play areas at the shelter. In contrast, older children tended to highlight the value of reliable Wi-Fi, staying connected with friends through social media, and engaging in outdoor activities such as cycling, going to the library or for an ice cream, or working on their mopeds. However, most children also talked about a lack of things to do, and being bored or isolated indoors due to the security measures interrupting their everyday lives. Thus, the rules and restrictions for safety and security at a DV shelter also have their downsides, restricting the children’s opportunity to engage in indoor and outdoor activities.

Social Relations: Friends, family, and Shelter Staff

For most children, the ability to stay in touch with friends and family is not only limited by safety precautions, but also related to the move from their old home to a new place, which is often distant from where they lived before. Their friends and staff at the shelters, or new friends in the town where they have moved, thus become very important for the children. Some children living in shelters with shared accommodation also make friends at the shelter, as Lilly (girl, age 7) explained: “You make many friends, who you live with [at the shelter]. [Name] and [name] are my friends.” Other residents are thus important, but the children also said that it is difficult to make friends when there might not be any children of the same age or with the same interests, as in Natasha’s case.

I can’t have anyone over, and I can’t really tell the truth directly either, so it was like just yes but, I don’t need to have friends, I can wait until high school. There was only one year left. But now I feel that it won’t work, that I need to have someone by my side, and I have a friend who’s at my school. But I want more […] so I wish someone my age would come [to the shelter] just so I could be close to her like that, so I’ve been looking out for that. But no one comes. […] Then I wouldn’t have had to explain everything to her. She would have understood me more, as it were. (Natasha, girl, age 14)

As Natasha’s narrative highlights, making new friends in a new place is not easy due to security precautions, meaning that you cannot say where you live, bring friends to the shelter, or tell them about your life situation on the run. Her narrative also reveals a longing for a friend of the same age to connect with and share experiences, something that had not happened.

For children who do not make friends at the DV shelter, or are unable to go to school or engage in activities outside the shelter, relationships become limited. A few children talked about only spending time with their mother and siblings, having no one else to socialize with. Elin’s narrative is an example of this.

So, it was quite boring not getting to meet so many people and such, but we used to still be able to go out shopping and things like that. And then you met other people, but then you didn’t talk. (Elin, girl, age 11)

As Elin’s narrative shows, the children’s social relations were sometimes very limited; they were unable to meet their old friends and family or develop new relationships in the new location. During the initial period at the shelter, due to security precautions, this seems to have been particularly difficult, and sometimes they hardly interacted with anyone except their mother and, if they had them, sibling/s. However, some of the children described how, after a while, they sometimes re-established contact with family, such as grandparents, as Jill’s (girl, age 8) narrative shows when we asked if she could meet anyone outside the shelter:

Jill: Yeah, we’ve met my grandma. But not the others.

Interviewer: And who are the others then? When you say not the others.

Jill: [Name of grandfather], grandfather’s girlfriend.

Interviewer: So, you’ve been able to meet some, but not all?

Mother: We haven’t been able to, no. Not grandfather. Only grandma. […]

Interviewer: How did it feel? Not being able to meet them.

Jill: It was sad that we couldn’t. We wished everyone could live in the same place in [name of city].

Thus, reconnecting with family outside of the shelter environment, after the initial time at the shelter, is very important. However, the required safety precautions make it difficult for children to stay in contact with family outside the shelter, and even more so with friends. It is rare for them to be able to meet physically; instead, they write or talk over the phone. Exceptions apply, however, to the few children who are able to stay in a DV shelter close to where they lived before. Although they must also deal with security risks, in some cases they can still go to school and see their old friends.

The children also spoke about the shelter staff, who appeared to play a significant role in their daily lives. Staff members were described as facilitating access to schooling and organizing activities, as illustrated by Mattias (boy, age 9).

Interviewer: You mentioned that you like the staff.

Mattias: They’re fun to be around and they’re kind.

Interviewer: Do they usually organize things to do?

Mattias: Yes, sometimes we grill sausages, and we had “waffle day,” then we made waffles, and we grill hamburgers too.

In addition to facilitating access to schooling, organizing activities, and, as illustrated by Mattias, being playful companions, the staff were often the individuals the children were most likely to turn to (apart from their mothers) when they had questions or needed support. This can be illustrated by Natacha’s answer, when we asked if there was anything she did not like about them moving away from the shelter:

That I lose contact with the staff. Because even though they haven’t been in my life for that long, they have kind of helped me the most, so it feels kind of sad to lose contact because I don’t want to. Because they have kind of helped me so much, so they’re a huge part of my life now. (Natacha, girl, age 14)

The children’s narratives suggest that shelter staff were important figures, although their roles were described differently depending on the child’s age. Younger children tended to describe staff as kind and playful companions, while older children emphasized the staff’s role as trusted adults from whom they could seek support. These differing perspectives highlight the multifaceted role of shelter staff in meeting the needs of children across age groups. However, there were also two children who expressed negative experiences regarding the staff, including Lisa (girl, age 15) who said: “There was a woman who worked there; she was really mean.”

There was also one person whom all the children said they did not want to meet, or had mixed feelings about—the violent offender, who was usually their father. A couple of the youngest children mentioned that they missed their father, while simultaneously worrying about or being afraid of meeting him. A couple of the older children had tried to stay in touch by sending mail, but still said that they did not want to meet physically.

To summarize, social relations were given importance in the children’s narratives, but were not always easy to sustain, often absent, and sometimes complicated or negative. The narratives illustrate the importance of being able to maintain contact with friends and family outside the DV shelter, and also having safe and supportive relationships at the shelter, with staff and other children. The narratives also show that a few of the children seldom met anyone except the staff and their mother and siblings.

Discussion

Generally, the themes we identified in the children’s narratives confirm previous research and contribute to the research field by finding that, in Sweden, as previously found in other countries, a stay at a DV shelter, although necessary for children’s safety, often comes at a cost. Specifically, the children in our study reported feeling safe within the shelter environment, which is consistent with earlier research (Arnell & Thunberg, 2025; Chanmugam, 2011; Øverlien, 2011a; Thunberg et al., 2022). However, their accounts did not portray the shelter as a “safe haven,” but rather, in the absence of meaningful activities, toys, opportunities for outdoor play, and basic prerequisites such as adequate space for the family and a clean living environment, the image they conveyed was more of a place of “safe storage.” These basic prerequisites, rarely emphasized in previous literature, emerged as critical for the children. Our study thus contributes to the field by foregrounding the importance of these basics, the absence of which has a negative impact on children’s quality of life during their stay at DV shelters.

The way in which the children interviewed in this study described the isolation, shelter rules, and restrictions on social relations that often accompany staying at a DV shelter is also in line with previous research (Chanmugam, 2011; Øverlien, 2012; Thunberg et al., 2022). The analysis reveals, for example, the importance of friends and their frequent lack, with the children often having no one of the same age to play with or talk to at the shelter. Still, our findings suggest that, while relationships with family members could often be sustained or re-established—with support from the mother through letters or phone calls—the children’s own (previous) friendships were frequently missed and negatively affected by the move to the shelter. It can be assumed that this is related to the security measures they lived with due to the threats of violence, and for some the long move to a place far away from their home. Thus, it is essential for their quality of life that children in DV shelters have access to structured leisure activities and schooling together with other children of their own age. Such interactions provide social interaction and peer support.

Furthermore, the narratives show that these children had different experiences and talked about both positive and negative aspects of the shelters they had stayed at. For example, when living in a DV shelter with shared accommodation, the opportunities to establish friendships and receive support from the shelter staff were valued. On the other hand, having their own apartment meant that they did not have to worry about how other children felt or acted, or share toys and spaces with other children. Thus, our findings reveal the importance of the shelter environment, activities, and social relations for children when staying there, and that their experiences can vary, depending on whether they have opportunities to play, spend time outdoors, or meet with friends and family. In line with the findings of Woodward and Joseph (2003), this suggests that, if shelters can enhance children’s experiences of security and safety, as well as their agency, control, and freedom in everyday life, the children will experience a better quality of life. At its best, this can lead to change, enhancing the liberation, growth, and wellbeing of children living in DV shelters.

The Process of a Shelter Stay

Although our empirical material is too limited to enable a structured analysis over time, one could hypothesize that a feeling of safety can start to develop more or less immediately upon arrival at the shelter, and that settling in by establishing basic requirements and discovering opportunities to play can also begin soon after arrival. Still, it can take some time to get used to the new situation at the shelter and build trusting relationships with other residents and staff (Thunberg & Arnell, 2025). Although it was not entirely clear in our interviews, we consider it likely that feeling bored or isolated develops later, depending on factors such as whether the available opportunities to engage in activities, play etc. offer the child enough stimulation. Similarly, it is possible that it is the move to the shelter that generates the interruption to social relationships and schooling, and that exploring the possibilities for social relationships with other children or with staff at the DV shelter, as well as homework, are later strategies to deal with these interruptions. If these strategies do not work out, children may suffer long-term distress, missing friends, and/or having to repeat semesters at school.

The process of a shelter stay also needs to be discussed in relation to the multiple stays that many children experience due to repeated threats. To understand the process of DV shelter life, we need to appreciate that every new move means that the children and their mothers have to start all over again, including arrival in a new town, not attending school for a while, meeting new staff at the shelter, and once again leaving toys and friends for a new and unfamiliar place. In line with Selvik and Øverlien (2015), we thus argue that a DV shelter stay can be understood as neither linear nor a temporary disruption; rather, it needs to be understood as a process of repeated interruptions affecting children’s quality of life. Furthermore, children’s quality of life and well-being involve something more than the mere absence of problems, such as violence, and typically focus on individuals’ subjective views of their own health and life circumstances. When discussing their quality of life, children include not only dimensions such as security and safety, but also agency and control in everyday life, freedom, fun, and physical health (Grip et al., 2014). In a similar vein, Woodward and Joseph (2003) suggest that having a sense of liberation, freedom, coping, and control are important for post-violence growth. It is thus important to ensure that the environment becomes as stable and predictable as possible for these children, supporting their safety, agency, and control (Grip et al., 2014; Hauge & Kiamanesh, 2019).

Limitations

The small number of interviews with children is a limitation on our ability to identify possible differences between children of different genders or ages. This is a common limitation in research with children at DV shelters. Despite this, the main strength of the study is that the interviews were conducted directly with a hard-to-reach sample of children, of different genders and various ages, instead of by-proxy interviews with their mothers or professionals. Their narratives enabled us to explore their experiences directly. Nevertheless, more research based on children’s own descriptions of their experiences is needed, from various countries and welfare systems, and this study has contributed to filling this knowledge gap to some extent. Moreover, it is also important to access the perspectives of the shelter staff and teachers, especially following the implementation of new legislation in Sweden, as well as the perspective of extended family members who lose contact due to the threat posed by the perpetrator.

Implications for Practice and Future Research

In relation to DV, both mothers and children have limited power to influence their situation. Not only are they the ones who must move, but the consequences extend further and affect the children’s quality of life. Adding a societal level to this discussion, it is obvious that the system of protection that has been developed globally has limitations when it comes to serving the rights and needs of children and their parents. Children living in DV shelters have the right not only to be protected from violence, but also to development and schooling etc. (Thunberg et al., 2022). The state must also implement all appropriate legislative, administrative, social, and educational measures to give each child and the people who take care of them the necessary support, according to the UN Convention on the Rights of the Child (United Nations, 1989).

Our finding that school presents a significant challenge for many children residing in DV shelters aligns with and extends previous research, such as that by Ekström and Hvenmark (2025). They found that the quality of shelters not only includes the relationship between mother and child, but also “that making sure that these children can also attend school while staying at a shelter constitutes a true quality measure” (2025, p. 7). Our study adds to this understanding by highlighting that children themselves express similar concerns. They underscore the importance of maintaining access to education during their stay, suggesting this is a critical area in need of improvement. Although our research is not comparative in nature, our findings also resonate with those of Øverlien and Selvik (2025), who show that children in DV shelters in Norway do not receive the education to which they have a right. Being denied access to education puts children in DV shelters at risk of falling behind academically, which could affect their long-term quality of life. Finding ways to solve this, for example by virtual schooling or schooling at the shelters if it is unsafe for children to attend in person, is of great need.

Some children in our study said that they missed their father, while at the same time worrying about or being afraid of meeting him. This highlights a particular challenge in navigating potentially divergent perspectives between mothers and children. While mothers residing in DV shelters often face conflicting demands—balancing the need to protect their children from a violent father with external pressures to maintain father–child contact (Andersson, 2024)—children, particularly younger ones, may be unaware of the associated risks and safety concerns. This underscores the importance of ensuring that both placement authorities and DV shelter staff are well-informed about each child’s individual circumstances, needs, and preferences, with their security as the number one priority. Mothers sometimes also feel compelled to leave their abusive partner or accept a shelter placement under the threat that their children may otherwise be taken into care (Andersson, 2024). In such cases, requirements for father–child contact can undermine the protective intent of the shelter stay and complicate the implementation of safety measures—particularly when visitation rights have been granted by the district court (Andersson, 2024). If a mother agrees to a shelter placement primarily to prevent the removal of her children, despite not believing it to be the most effective means of avoiding violence, this too must be acknowledged by those responsible for placements and by shelter staff. It is crucial to recognize these complex dynamics when designing interventions that genuinely prioritize the safety and quality of life of both mothers and children.

Based on our material, we suggest that shelters need to be better adapted to meet children’s needs, both when they arrive and while they are living at the shelter. Thus, the different routines and methods used for supporting children developed by shelters should be evaluated so that examples of good practice can be used nationally. We want to stress that DV shelters are not a problem per se; they are a solution to men’s violence. Whether DV shelters are the best solution is another issue, as children are generally not only removed from violence, but also from their friends, school, and extended family. We need to find ways for children to be protected from DV without being forced to move away from their home and losing their familiar living and school environment, their teachers, friends, and extended family members, their own toys and leisure activities. To make this possible, we suggest that policymakers investigate and find ways to protect children from DV by addressing and moving the perpetrator, rather than uprooting the children from their home. This must, however, be under the condition that their safety can be guaranteed (see e.g., Diemer et al., 2017), and that it comes without costs to other dimensions of the children’s quality of life.