‘They Yell and I Yell Back’ Pre-schoolers’ Descriptions of Conflict Laden Interactions at Home

Over the past 50 years, extensive research has been published on the parent–child relationship and parenting. However, there are very few examples where young children are at the centre of attention for describing family dynamics, relationships and conflicts. This study aimed at addressing this research and knowledge gap through exploring the emotional and relational experiences of preschool children whose parents attended a universal parenting programme. Seventeen preschool children aged 3–6 were interviewed, using an emotion-focused, pictorial-based computer assisted interview method. The children’s descriptions of their family relationships were analysed using qualitative content analysis. The children described negative interplay within the families in rich detail, especially experiences where conflicts with parents escalated and were left unresolved. Moments of positive family interactions were described as well, but they were heavily overshadowed by the narratives containing negative parenting. The children also described compensatory behaviours, such as looking for comfort from siblings or pets. The narratives in this study gave a unique insight into the emotional and relational domestic context of children in families seeking universally offered parenting support. Given adequate tools and support, children as young as 3 or 4 years old could provide extensive information about their lives. We urge future research examining parenting or family interventions to include the children’s perspectives.

The preschool age is a period in life characterised by rapid development in several areas.Gross and fine motor proficiencies are improving fast (Adolph, 2002;Butcher & Eaton, 1989), they improve in mastering the complex art of language and communication (Conti-Ramsden & Durkin, 2012;McCarthy, 1943), as well as developing social skills and cognitive abilities.This development is not uniform and there is a wide variability in proficiencies.For example, while most children have developed the use of multiclause utterances, turn-taking and other rather advanced language related skills at the age of five, the variation is considerable (Conti-Ramsden & Durkin, 2012).To identify, express and regulate emotions is also an important capacity that develops throughout childhood, in close interplay with the caregivers and coincidentally with the developing frontal lobes (Riediger & Bellingtier, 2022).Additionally, the preschool age is a period when conflicts can start to emerge in the child-parent relationship.The child is testing limits and exploring his or her own autonomy and competence.Parents, especially first-time parents, are developing their own parenting skills and also need to adapt to new challenges and be responsive to their child's developmental changes.Although extensive research has been published on parent-child relationship and parenting during the last 50 years, there are very few examples where preschool children's own views are the focus of analysis when describing family dynamics, relationships and conflicts.

The Child-parent Relationship
A family can be described as a dynamic system, wherein stable behavioural patterns are shaped through interactions between members of the family (Granic & Patterson, 2006;Lunkenheimer et al., 2016).The family members make up the foundation of the system, as they are interconnected parts that affect each other over time.The definition of a family system depends on area of research or research question (van Geert, 2019): it might be the entire family or dyads within the family, e.g., child-parent.For this study, we will be focusing mainly on the child-parent dyad.
The transactional model posits that an individual's developmental processes are shaped by the reciprocal interactions between the individual and their surrounding context over time (Sameroff, 2009).In this model, a child's development is the result of ongoing and dynamic interactions between the child and their social environment, especially the parents, leading to patterns in reciprocal interactions over time.Both positive and negative interactions (such as fighting, yelling and escalating conflicts) become constant parts of the interaction between parent and child in what can be described as a coercive cycle (Patterson, 2002).What is noticeable about child-parent relations in a coercive cycle is that they are simultaneously rigid and inconsistent (Lunkenheimer et al., 2016).The rigidity is displayed through parents and children being stuck in predictable, negative patterns of interaction, while the inconsistency is displayed through parents' inconsistent or unpredictable emotional and behavioural responses.These malfunctioning, negative interactive patterns are often preserved and are likely to reoccur in future child-parent interactions (Granic & Patterson, 2006).

Family Conflict and Parenting
In many families, conflict is part of everyday life and for some children family conflicts are a normal part of life.Conflicts between parent and child are associated with several factors, notably parenting style (Chang et al., 2003;Krishnakumar & Buehler, 2000), but also inter-parental conflict (Gerard et al., 2006) and child temperament (Rubin et al., 2003).These conflicts are described by children as associated with negative emotions such as sadness, anger and embarrassment (Nixon & Halpenny, 2010).Researchers have identified two domains of parenting: parenting practice (e.g., parental monitoring, involvement, and goals) and parenting style (the emotional conditions in which the child is raised) (Spera, 2005).Parenting style, especially harsh parenting, is linked to higher levels of family conflicts and behavioural problems.Growing up in a family characterised by intense conflict and harsh parenting is associated with increased child behavioural and relational problems in both the short and long term (Bayer et al., 2012;Jaycox & Repetti, 1993;Klahr et al., 2011;Lunkenheimer et al., 2017).Since 2015, mental health problems, especially behavioural and emotional problems, are considered the second highest cause of burden of disease in Europe and the Americas (Baranne & Falissard, 2018).Dysfunctional parenting, especially harsh parenting, is associated with risk of increased relational aggression in children (Kawabata et al., 2011).Conversely, Kawabata and colleagues found that positive parenting behaviours, such as providing warmth, help and positive reinforcement, are associated with less relational aggression in children.This association between harsh parenting and child behaviour problems seems to be bidirectional to some extent, indicating a possible transactional dynamic at play where child and parent are influencing each other's behaviour (Lansford et al., 2018;Shaffer et al., 2013).Parenting programmes are generally the recommended interventions for addressing said problems (Furlong et al., 2012;Socialdepartementet, 2018), and are the main reasons for seeking parenting support in Sweden (Thorslund et al., 2019;Wells et al., 2016) Children's Views on Family Relations Despite extensive research on parents and children, children's perspectives have rarely been studied.This is problematic, as children's experiences are important for various reasons.One reason is the unique insight and information children can provide on their own experiences, well-being, and opinions.Information that can be important and sometimes crucial for providing the best intervention to meet their needs (Cross & Hershkowitz, 2017;Wilson et al., 2010).Children's own perspectives could also be important when tailoring and evaluating interventions targeting them and or their families.Another reason for obtaining children's perspectives is children's rights to be heard in all matters concerning them and their right for information, as described in the United Nations' convention on the rights of the child (UNICEF, 1989).However, systematically involving children in practice as well as in research has been widely ignored or overlooked in the past.During the last decade, the importance of children's voices has been stated both in research and for evaluating interventions (Baird & Grace, 2017;Molloy et al., 2019;Nilsson et al., 2015).Children's perspectives have been actively sought in such disparate research areas as healthcare (Schalkers et al., 2015), preschools (Almqvist & Almqvist, 2015), and forensic contexts (Baugerud et al., 2020).Despite this, research on parent-child conflicts and parenting often focuses on adolescents or pre-adolescents (Strazdins et al., 2017), or on families with specific or substantial challenges, such as cancer (Darcy et al., 2019) and parental divorce (Hayes & Birnbaum, 2020).To our knowledge, there are no studies involving the views of young children in families seeking parenting support.
In previous studies where the child perspective has been highlighted, only a few researchers assess the child's opinions and perceptions of family relations (e.g., Bost et al., 1998;Rooth et al., 2020), while most tend to rely on hypothetical stories for the child to discuss or by letting the child develop their own stories using figures, which are then analysed (Shamir et al., 2001;Yamada, 2009).When research has focused on children with suspected emotional or behavioural problems, obtaining the child's own perspective has traditionally not been prioritised, due to somewhat misinformed beliefs of children not being capable of reporting their own difficulties (Edelbrock et al., 1985).These beliefs are not supported by research on child development.On the contrary, studies show that children from the age of 2 can describe memories of personal events (Peterson, 2002(Peterson, , 2012)), including emotions using terms such as happy and sad (Fivush & Baker-Ward, 2005).The reports of young children are often brief and sparse, however, the narrative skills and memory retrieving skills and capacities are constantly developing (Nelson & Fivush, 2004;Schwenck et al., 2009).When comparing children aged 5 and 6 to children aged 3 and 4, the older children were more capable of providing informative responses and details compared to the younger (Hershkowitz et al., 2012).However, across children of the same age there is great individual variation in the ability to describe their experiences.These are affected by, among other things, differences in the caregiver's language and reminiscing style, children's cultural differences, and family context.In addition, children with behavioural and emotional problems often exhibit speech and language difficulties, which in turn affect their ability to produce a verbal narrative.To summarize, children's own perspectives have historically not been included in research on parents and families.Today there is still a knowledge gap regarding children's own perspectives and opinions on their family relations, especially with respect to the younger children living in families with conflicts.

Objectives
The current study aims to address this knowledge gap on children's perspectives by examining preschool aged children's views on their family relations.In particular, our objective was to explore the expressed emotional and relational experiences of children aged 3-6 years whose parents attended a parenting programme.This study was conducted within the context of evaluating a parenting programme, offered to parents of preschool-aged children with perceived emotional and/or behavioural problems.

Research Design
Qualitative content analysis was applied to describe the children's descriptions and perceptions of their family relationships with an inductive approach (Graneheim & Lundman, 2004).The study was approved by the Regional Ethical board in Uppsala, Sweden (dnr.2018/188).

Participants and Context
Purposive sampling was applied to recruit participants among families entering a Triple P group parenting programme in Uppsala, Sweden.Triple P is a parenting programme based on social learning theory, originating from Australia.It has four main foci: (1) to increase the parent's competence through building a positive parent-child relationship, promoting positive and prosocial behaviours, and managing common emotional and behavioural problems; (2) to reduce the use of coercive and punitive parenting behaviours; (3) to help parents communicate about parenting issues; and (4) to reduce parental stress (Sanders, 2012).In Sweden, Triple P is available as brief seminars, individual counselling, and as parent groups.For this study, families participating in the group intervention were recruited.
The parenting programme was delivered at preschools, by preschool teachers, and offered universally to all parents of preschool children within the municipality.No inclusion or exclusion criteria were applied.For more information about the evaluation of the implementation of the programme, please see Dahlberg et al. (2022).Parents of 21 out of 24 children eligible for interviews consented to participate in the study.One child declined participation and the parents of three children withdrew their consent, resulting in 17 interviewed children, aged 3-6 years, interviewed at 7 different Triple P groups.All interviews were conducted by the first and last authors, both trained child psychologists.Mean age was 4 years (7 three-year-olds, 5 four-year-olds, 3 five-year-olds, and 2 six-year-olds).Gender distribution was quite uneven, with 6 girls and 11 boys participating in the interviews.Four of the children had foreign-born parents.No further data were collected on characteristics of the participants.However, in a related study, data from 125 families having attended the programme previously were available, assessing child emotional and behavioural problems (Strengths and Difficulties Questionnaire, Eyberg Child Behavior Inventory), parenting practices (Parenting Scale), and parent mental health (General Health Questionnaire).Analyses of these questionnaires revealed that half of the children scored above clinical cut-off on the Strengths and Difficulties Questionnaire and almost 3/10 on Eyberg Child Behavior Inventory, while more than 2/5 of the parents scored above cut-off on the General Health Questionnaire and 3/10 scored more than one standard deviation above population mean on the Parenting Scale (Dahlberg et al., 2022).This indicates that the children in the present study were more likely to come from families at risk, burdened by mental health-related problems to a higher extent than the general population and might be at greater risk of experiencing negative parenting.However, it's important to note that similar data for the children in the present study were not available.
The language proficiency of the interviewed children varied greatly.While some children struggled to form words and build sentences, other children were able to use long and interconnected sentences and statements.Further, and as a side note to this study, the three temperament types described by Thomas and Chess were recognisable during the interviews: while some children were shy and withdrawn, others slow to warm up, yet other children were apparently comfortable from the get-go and displayed a general positive mood (i.e., difficult, slow-to-warm-up, easy) (Chess & Thomas, 1977;Thomas & Chess, 1977).

Procedure
Data collection took place from November 2018 to March 2020.In connection with an information letter about the Triple P parenting programme, a separate invitation to participate in the study was sent out to parents of children 3-6 years old from seven selected Triple P groups, covering geographic areas and facilities with different demographic make-up regarding parental education level, cultural background, as well as urban and rural populations.The invitation letter contained information about the study, both information directed at parents and a separate information letter to the children.Informed consent forms were distributed on site to parents who accepted the invitation to participate in the study.The researchers also provided the study information at Triple P group start-up.Assent was sought for from all children at the beginning of the interviews.One child, whose parents consented to participate, declined participation.The interview length ranged from 12 to 30 min, with a mean time of 17 min.Children received a movie ticket and a choice of rubon tattoos after participation.
All interviews opened with the researcher presenting themselves, followed by seeking assent and outlining the interview procedure.Four ground rules were established and explained by the interviewer: "If I ask or say something that you don't understand, you can say 'I don't understand'"; "If I say something that is wrong, you can tell me"; "If I ask something and you don't know the answer, you can tell me by saying 'I don't know'"; "If you want to quit, anytime, you can just say 'I don't want to continue' or show me like this for example [signals "stop" with open hand]".The first three ground rules are known to enhance the accuracy of children's recall (Hershkowitz et al., 2012), while the fourth rule is to ensure children's voluntary participation.Five children chose to verbally or nonverbally end the interviews prematurely, although all these interviews were at their closing stage at that point.
The computer-assisted interviewing aid In My Shoes (IMS) was used as a structure for all interviews (Calam et al., 2000).IMS is a software developed specifically towards younger children and children with communication difficulties, and has proven to be effective in eliciting both accurate and elaborate accounts from children, without jeopardising one at the expense of the other (Barrow & Hannah, 2012;Bokström et al., 2016;Cousins & Simmonds, 2011;Fängström et al., 2016;Fängström & Eriksson, 2020).Additionally, IMS is a helpful tool in aiding children with language disabilities or communication difficulties to express themselves (Cousins & Simmonds, 2011).Via IMS, children are assisted to visually represent their families and relations therein, and to give accounts of their own and other people's behaviours and emotional experiences.The software works as a visual aid and as a starting point for the child to narrate.
There was a focus on the child's family relations in the interviews, which was supported by letting the children name and choose a figure that represented each member in their family.Further, the children's narratives were elicited through asking about experiences related to different emotions.At the beginning of the interviews, children were asked to name different emotions in IMS, based on standardised smiley faces (see Fig. 1 for an example of labels).These emotion labels were used throughout each interview as a conversational starting point to obtain the child's descriptions of themselves and other family members' behaviours and emotional states.The interviewer alternated between negative and positive emotions in a specific order, to ensure that emotions were not covered from "good" to "bad", or vice versa.In addition to using the IMS software as an interview aid, the interviewers ensured to adopt best practice for child interviews as proposed by National Children's Advocacy Centre (The National Children's Advocacy Center, 2012), using open questions and providing a safe environment for

Happy
Ordinary A bit sad Sad Angry Scared Fig. 1 Emotions palette used in the interviews, with an example of a child's own emotion labels the child to speak in.Interviews were continually carried out until a saturation point was reached.

Analysis
The interaction between interviewer and the child was audio-and video-recorded, as well as activities on the computer screen.The interviews were transcribed verbatim by external transcribers.Because of the scarcity of research related to the present study's topic, qualitative content analysis was applied to analyse the data (Graneheim & Lundman, 2004), using an inductive approach (Elo & Kyngäs, 2008).Initially, all interviews were reviewed by two of the authors (AD, KF) to acquire an overall sense of the material.Both transcriptions and original recordings were examined repeatedly.Subsequently, the manifest content was condensed into units containing the main propositional content, labelled with codes.Next, all codes were abstracted and inductively sorted into categories based on their commonalities.Since distinct patterns within the categories were recognised during this part of the analysis, the categories were divided into sub-categories.Both categorisation and sub-categorisation were attempted and discussed until agreement between all three authors was reached.See Fig. 2. for an example of the analytical categorisation process.In a final step of the analysis procedure, an overarching theme was conceptualised from the categories.

Results
The interviews had a clear focus on emotional and relational aspects of children's experiences in relation to their family members.A map of a theme, categories, and subcategories are presented in Fig. 3 above.The following categories were identified in the analysis: Negative child-parent interactions, Lack of positive child-parent interactions, and Positive and secure family interactions.Each category consisted of three to four subcategories.Excerpts from the interviews are presented as single quotes or dialogue, to illustrate subcategories.Based on the subcategories and categories, one overarching theme emerged, presented at the end of the Results section.All names were pseudonymised and other potentially identifying information has been obscured.

Negative Parent-child Interactions
This category consisted of three sub-categories: The quarrel starts, They yell and I yell back, and No resolution, everyone is sad.The category was characterised by the presence of negative and dysfunctional interplay between child and parent.When put together, the sub-categories constituted parts of what resembles the coercive cycle.The coercive cycle encompasses how members of a family, often the parents and children, get stuck in negative relational interactions.The children's narratives were rich with descriptions of their own and their parents' and sometimes siblings' negative emotional reactions and relational behaviour.Furthermore, the narratives also included temporal aspects in that children included and gave examples of the important building blocks of the coercive cycle, i.e., the start of the conflict, the escalation trap and the ending of conflicts without a resolution.The relational patterns in this category included strong negative emotions of the child, the parents and sometimes the entire family.

The quarrel starts
Children described various situations that constituted the start of a conflict between them and their parent/parents.When parents for example asked children to end what they were doing, such as watching tablet or TV, it evoked anger and frustration in the child.One child described that he watched TV and didn't want to end that and leave home: I was watching TV… And then mo-dad came and said that we had to go.Then w-then I became angry (Calvin, 5 years old) Another situation that several children described as making them angry, was when their siblings teased them and this Quote I was watching TV.Mhm.And then m-dad came and said that we had to go and then I got angry.

Angry because must stop activity
Sub-category The quarrel starts

Category
Negative child-parent interaction R: I: R: Fig. 2 Example of the analytical process, from quote to category was often the start of a more complex conflict situation involving the child and both parents.At other occasions the children described conflicts in the family associated with feeling sad.These situations arose for example when a sibling hit or threw something at the child, which in turn made the parent angry.Conflicts that evoked sadness could also arise when a parent was stressed and, for example, forced the child to hurry up when going to preschool.
Another time when I go to preschool, I was sad on the way home because I want-I didn't want to go so fast.Mo-mom said we can walk fast because we were late b-but I don't want to mom.I… mom walks too fast but mom pushed me because I didn't want to (Natasha, 4 years old)

They yell and I yell back
The children provided rich descriptions of what can be understood as escalation of the conflicts.Overall children described their parent or parents as becoming angry with them and/or their siblings as part of a conflict situation.Parents both looked angry and acted in an angry manner.
And sometimes I can feel angry and then I slam the door and it goes 'bonk'[…] They come and yell at me: 'Don't slam the doors!' (Adam, 6 years old) Children often described that they reacted with yelling back or throwing things.Some children also slammed their door, as the child in the quote above.The descriptions of parents becoming angry existed irrespective of the emotional state of the child (feeling angry or sad).However, children's descriptions of situations when they felt sad was mainly related to the angry parent acting in a coercive manner.Children gave examples of their parent's threatening them with punishment or physically forcing them to do things such as staying in their room.
Mom says 'Yy-you will get a punishment'.Eh, eh, I say 'I will not get a punishment' (Russell, 4 years old) A number of children described that their mothers and/or fathers hit them and their siblings: She gets angry and sad when I do mean things to her, then she gets angry […] Then she hits me (Leon, 4 years old) When parents acted in a coercive manner, children reacted with protesting verbally or physically, such saying "No" or locking themselves in their room and not letting the parent in.Children also described trying to hide or to run away: And dad he-he-he couldn't… he hi-he not hit us, not me and John.Then-then-then we were quick (Josef, 3 years old)

No resolution, everyone is sad
The conflict situations children described that made them angry or sad did not contain any descriptions of conflict resolutions, e.g., parental behaviours to compromise or reconcile with the child.The most common account that described the end of these situations were that the parent/ parents were sad or everyone was sad: Uhm, then, then, then dad hi-then mom was sad and John was angry and I was sad, too […] A-and dad was sad, too (Josef, 3 years old)

Lack of Positive Child-parent Interactions
This category consisted of three sub-categories: Mom watches TV and dad is working, My siblings are my helpers, and I want everyone to be happy.The category was characterised by an absence of statements and descriptions of positive or neutral child-parent interactions.This was a pattern that clearly emerged when all the children's narratives were reviewed together.In the void of positive child-parent interactions, siblings and pets seemed to have an important and compensatory function as helpers and someone to go to for comfort.The relations with siblings and pets are where the children find happiness.In addition, children expressed a desire for everyone to be happy, which could be understood as a longing for a positive and stable state for the family.

Mom watches TV and dad is working
In some narratives the parents were absent from all accounts the children made of their life at home.In other interviews children did not describe any positive situation or activity that involved their parent.In these instances, children could describe that their parents did things such as working, watching TV or sleeping.But they did not provide any narratives of them doing things together.The following quote is from a child describing feeling happy when the family were all at home.The interviewer asked what they did and the child replied: Then, then mom watches grown-ups' TV.And dad, Isits and I and John dance and dad starts Playdance and… and dad is working (Josef, 3 years old) When parents were absent from the narrative children would instead describe for example feeling happy when being alone, when playing with friends, or doing things with their siblings.

My siblings are my helpers
Related to the absence of parents in children's narratives included in the subcategory Mom watches TV and dad is working, are children's descriptions of how they turned to their siblings or pet for comfort and help.One child described that she, when she felt "in between", was together with her siblings and the cat: My older sister and my baby and my cat… 'Cause they are actually my comforters… They comfort me (Heather, 4 years old) The same child also referred to her siblings as being her "helpers", providing support and help when she needed it.Children also expressed that siblings could help them feel better in a conflict situation with the parents: Respondent: He-he-uhm, he-he made me laugh… Yes, he tickled my feet Interviewer: Yes, and how did that make you feel?Respondent: Happy (Calvin, 5 years old) I want everyone to be happy Some children expressed that they wanted everyone in the family to be happy.These statements were not formulated as if the members of the family were happy, but rather as a wish that it were so.In the following excerpt, the child points at the family members at the computer screen and expresses that they should all be happy: Respondent: But, but that one too, and that one [points to the screen at the mother-figure] Interviewer: Is mom with you when you are happy?Respondent: Yeah Interviewer: What do you do then, when you are happy and at home?Respondent: I want that one to be happy too, and that one is happy [points at members of the family] […] and that one happy and that one happy [points at members of the family] (Casey, 4 years old) These statements were often combined with an absence of examples of positive interactions with family members.

Positive and Secure Family Interactions
This category consisted of four sub-categories: Everyone plays together, We have a cosy Friday night in, I turn to my parents for comfort, and Conflict resolution.The positive interactions encompassed doing things together, like playing or creating, as well as being close to each other in a more physical and cosy way.Children also described how their parents comforted them when they were sad or angry, either from hurting themselves or after sibling conflicts.Further, conflicts between parents and children were resolved.This category was characterised by a sense of security in the child-parent relation, and thus named accordingly.
What was significant within these sub-categories was the short descriptions that not only lacked details, but also lacked examples of any interactive interplay, as well as temporal descriptions.In that sense, these statements were more like generic descriptions than vivid and colourful narratives.

Everyone plays together
The children described that playing together with the entire family made them happy.They gave examples of laughing together or tickling a parent.When compared to the narratives related to feeling sad or angry, the experiences associated with feeling happy were considerably less detailed.One child described being with the mother and father when feeling happy.The interviewer asked what they did then and the child replied: Mmm, doing good stuff.And that was all (Yasmin, 5 years old)

We have a cosy Friday night in
To do nice things together with their parents was described by the children as making them happy.Some examples were to paint or to ride a bike together or to talk about things that interested them.One child also gave an example of feeling happy when watching the parents dance with each other and having a good time.In addition, the Swedish tradition of having a cosy Friday night in (spending time together with the whole family), was also described as a good time with the parents and the family.
Several children described how they liked to cuddle with or be physically close to one or both parents and that this made them feel happy.
Mom will kiss me, just mmumumum (Adam, 6 years old) One child described the parent as making special food for him and always listening to him, something which could be understood as the parent being considerate.

I turn to my parents for comfort
Some children described that they would turn to their parents for comfort when feeling sad.This was especially common when a sibling had done something that hurt the child, such as hitting them or teasing them.
Someone [points at father on computer screen] comforts me if I am really grumpy with my brother… When he teases me (Kaitlyn, 3 years old)

Conflict resolution
In a few instances, children could give examples of conflicts not ending with everyone being sad, as in previous sub-category No resolution, everyone is sad.Instead, they described that the conflict ended with the child and parents cuddling and becoming friends again.OK, at the end of the fight we actually get happy, the both of us, and cuddle!(Yasmin, 5 years old) Emerging Theme: "Conflict Laden Interaction between Children and Parents with Moments of Happiness" Abstracting the categories encompassing the children's narratives, a picture unfolded portraying interactional patterns where coercion and conflict overshadowed play, tenderness and closeness.This resulted in the creation of an overarching theme named "Conflict laden interaction between children and parents with moments of happiness".The children's narratives particularly emphasised the presence and potency of negative interplay between child and parent while giving hints of other modes of interaction that the children longed for.
Within this theme, two distinct modes of interaction appeared, informed by the four categories.One mode described how children and their parents got stuck in coercive cycles, constituted by escalation, fighting that sometimes led to violence and a lack of conflict resolution, with the child seeking positive interactions and comfort from peers or siblings while longing for everyone to be happy.Instead of feeling happy, the children expressed unresolved feelings of sadness and sometimes anger.This mode also contained a temporal aspect, in the sense that the children described how things progressed within the coercive cycle and which event preceded the other.The other mode gave glimpses of interactions characterised by play and closeness.This was not necessarily a mode where "all is good", but rather a healthy climate for positive interactions to evolve and for conflicts to be resolved.Strikingly, this mode contained significantly less information: children's narratives where both scarcer and less detailed.Children were much more prone to give elaborate answers when discussing negative rather than positive sides of the family system.

Main Findings
The aim of the current paper was to explore the emotional and relational experiences of children aged 3-6 years whose parents attended a parenting programme.To our knowledge, this is the first study where pre-schoolers are informants as representants of parenting programme participants.The findings indicated the existence of several instances that described negative child-parent interactions, where children and their parents got stuck in coercive cycles.These narratives were marked by children's unresolved sadness and anger.The results also contained glimpses of a healthy climate for positive interactions to evolve and for conflicts to be resolved.However, these positive family interactions were both scarcer and less detailed.

Children's Descriptions of Family Relational Patterns
The children in this study described both parenting practices and parenting styles.The narratives described practices such as parental involvement and lack of involvement, which relates to parenting practices.But more prominent were the descriptions of parenting styles, especially harsh parenting.These negative descriptions of the emotional environment provide an extensive view into the children's everyday lives and the child-parent relationship on a system level.The narratives illustrate the consolidation of coercion similarly to Granic and Patterson's (2006) description of being "drawn by some invisible force beyond our control to repeat the same type of interaction".
It is well known that parents should focus on reinforcing positive interactions and stop coercive or harsh parenting strategies (e.g., Beauchaine et al., 2005;Galboda-Liyanage et al., 2003;Gardner et al., 2006;Hart & Risley, 1995).The results from the present study, however, suggest that the children's narratives reveal a context where the opposite might be true: negative child-parent interactions were the dominant mode in the children's lives.Not only negative interactions, but situations where the parents lost control and used violence were described by the children.Children with perceived behavioural problems are at greater risk of experiencing violence or coercive parenting at home (Bayer et al., 2012;Hipwell et al., 2008;Lunkenheimer et al., 2017;Rodriguez, 2010), something that the narratives from the interviewed children give voice to.The Triple P practitioners have a plan for addressing suspected domestic abuse when this emerges from the contact with attending parents, which includes reporting the families to the social services.However, no such reports or other measures were taken during the study period, which might imply that this information was unknown to the practitioners (personal communication with the local Triple P coordinator).This underscores the importance of getting the child's perspective, not only for providing a different view of the family, but as a source of information for discovering dysfunctional family interactions and child abuse.
The children also described how they compensated the lack of positive interactions with their parents by seeking support and security outside this dyad, through siblings, friends and pets.Siblings' unique role in promoting child development and compensating for dysfunctional child-parent relations has been explored in previous research, as well as the positive impact of pets on child development (e.g., Azmitia & Hesser, 1993;Melson, 2003;Milevsky & Levitt, 2005;Triebenbacher, 1998).
Although not within the scope of this study, it would be interesting to evaluate similarities and differences in children's and parents' descriptions of the relational patterns within the family.For school-aged children and above, there is evidence suggesting that children and parents describe parenting behaviours similarly, albeit with children describing parenting behaviours slightly less favourably than their parents (Korelitz & Garber, 2016).

Child Involvement
As described above, the importance of first-hand accounts from children in families that are subject to interventions from the society has been stressed in previous research.However, parenting support and interventions aimed toward young children have largely been developed and evaluated using parents and other significant adults as references.The results in the present study show that even very young children can provide highly important information on their experiences, relationships and well-being.The children's narratives could be useful in studying the mechanisms of the entrenchment of negative interaction within the family, and by extension the strategies necessary for change.Thus, they could serve as a part of the information-collection process that is the start of any intervention towards families.Furthermore, children's narratives could also serve as a motive for the parents to change their parenting practices.As relational patterns become more stable over time, early interventions are important for developing positive interactions and addressing negative ones (Guttentag et al., 2014;Mihelic et al., 2017;Sanders et al., 2000).This calls for methods of identification and involvement of children at an early stage and early age.
Interviewing children aged 3-6 is uncommon throughout research, a finding that also applies to research on parenting.Although the concept of a coercive cycle has been studied in great detail, and most social learning theory-based parenting programmes are informed by this concept, there is a dearth of studies where the child's own experiences of such patterns are explored.The present study adds to the already existing scientific knowledge of suboptimal relational patterns in family systems by providing the children's perspectives.
In order to include young children in practice and in research, with the aim of capturing their experiences, thoughts, emotions and opinions, there is a need for methods that take developmental level and child perspectives into account.The IMS computer-assisted interview, which was used in the current study, is a tool developed to accommodate young children's needs in an interview situation.IMS works as a structured visual aid, with pictorial representations of emotions, people, speech and thoughts and the software is used jointly by the child and the interviewer.This format is known to increase children's information sharing (Hamond & Fivush, 1991;Peterson et al., 2013;Saywitz & Camparo, 2013) and to ease their anxiety and stress in interview situations (Calam et al., 2000).IMS can achieve this without comprising the accuracy of children's recall (Fängström et al., 2016).The structure of IMS, which includes a systematic exploration of children's emotions and associated relations and experiences, is likely to have been helpful to the children when providing their perspectives.Previous research has demonstrated that IMS is to prefer over methods that solely rely on verbal information sharing, when the aim is to gain insight into children's emotional experiences (Fängström & Eriksson, 2020).

Study Context
Using an interview guide and an interview method validated for the specific age group of this study and the systematic analysis of data, contributed to the credibility of our results.Further, we attempted to describe the analysis process as thoroughly as possible, which involved all authors in different phases of the process, to facilitate future replications of the study.This study included 17 children, which calls for caution when transferring the results outside of the specific context of the study.Although the parenting programme from which the interviewed children were recruited is open for all parents to attend, analyses of questionnaire data suggest that participating families have higher levels of mental health related problems (Dahlberg et al., 2022).There is a vast body of work from research across the past decades suggesting that parental mental health, such as stress, is closely related to child behaviour problems and family conflict (e.g., Baker et al., 2002;Jones et al., 2021;Neece et al., 2012;Smith, 2004).This limits the level of transferability as the children in this study are likely to be more representative to a clinical sample than to the general population.Additionally, there are no available studies within the same context to reflect and compare our results with.However, families where the coercive cycle is pulling parents and children into repeated negative interactions is by no means unique to the participants in this study, which can make the results relevant in other contexts.Nonetheless, more interview studies are needed where preschool children are respondents, to assess whether this study's conclusions are applicable in other contexts.
Since no data were collected regarding the parents' view of their child's emotional and behavioural functioning, we do not know if they are representative of children of parents seeking general parenting support.However, recruitment took place in preschools from areas with a wide array of demographic make-up, such as parental education, general income, and country of origin as well as rural and urban areas.

Ethical Considerations
Following Swedish legislature, families where children are suspected to suffer from abuse must be reported to the social services.In this study, close contact was kept with the social services from the initialisation of the project, where the researchers could consult the social services where the child was deemed to be at risk.In some cases, notification to the social services was necessary when domestic abuse was suspected and the family needed the right level of support.

Conclusions
The narratives in this study gave a unique insight into the emotional and relational home context of children in families with high levels of conflict.The children described negative interplay within the families in rich detail, especially experiences of the coercive cycle and how conflicts escalated and were left unresolved.Given adequate tools and support, children as young as 3 or 4 years old could provide extensive information about their lives.We argue that the children's perspectives must not be omitted if one seeks to truly obtain complete information of the family and the relational patterns therein.We urge future research examining parenting or family interventions to include the children's perspectives.Such perspectives might even be used as part of the therapeutic process: presenting children's experiences and worldview to their parents might help parents implement the behavioural changes needed to exit the coercive cycle and increase positive family interactions.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material.If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/.

Fig. 3
Fig. 3 Map of data categorisation

Interviewer:
When you are at home, do you ever feel like this?That you are happy?Respondent: [nods] Interviewer: Mhm, tell me, who's with you?Respondent: Dad.Interviewer: Dad.And what do you do?Respondent: I don't know.Interviewer: When your happy?Wh-Respondent: Drawing (Hamzah, 3 years old)