Sage Journals: Discover world-class research

Abstract

Summary

Intense conflicts are prevalent in families receiving child welfare services. Research in the field of family interventions usually focusses on dyadic conflicts within the limits of the nuclear family unit, ignoring the larger social context of families. Therefore, this study analyses effects on interpersonal conflicts in personal networks of parents during a family intervention. The theoretical framework is based on the family systems theory, which highlights the role of mutual trust for conflict resolution. This research applied ego-centric network analysis in combination with a longitudinal design. The sample consists of 46 parents who were enrolled in a home-based family intervention in Switzerland.

Findings

The results show that development of mutual trust between practitioners and the network members of parents is associated with significant decrease in conflict density and conflict reciprocity. Hence, network members experience fewer interactions that are associated with conflict as well as less intense conflicts in the medium term. However, no effects were found on the centrality of parents in interpersonal conflicts.

Applications

The results of this study suggest that practitioners in family interventions must establish mutual trust with not only parents but also other individuals to promote a less stressful family environment. This emphasizes the importance of a network perspective on interpersonal conflicts in child welfare interventions.

Keywords

Social work family interventions child and family welfare conflict family support quantitative research

Introduction

Family conflicts are a common risk factor for the well-being of children who are recipients of child welfare services (Autorengruppe Kinder- und Jugendhilfestatistik, 2021, p. 22). Interpersonal conflicts in families have wide ranging negative effects not only on parenting but also children's well-being (Grych & Fincham, 2001; Sarrazin & Cyr, 2007). For instance, interparental conflicts are associated with adverse parenting such as harsh punishments or low parental acceptance (Krishnakumar & Buehler, 2000) as well as negative outcomes in numerous domains of child functioning, such as externalizing (e.g., aggression, delinquency) and internalizing behavior (e.g., anxiety, depressive symptoms) or emotional, cognitive, behavioral, and physical responses (van Eldik et al., 2020).

In most studies regarding family intervention outcomes, conflicts are either operationalized in a dyadic form (e.g., parent–child, parent–parent) or, in fewer instances, as family conflicts, but without any specification of the family unit. In both cases, however, conflicts are implicitly limited to the nuclear family unit. Current family research emphasizes the fact that dyadic interactions are influenced by a larger web of interdependent relationships (Widmer, 2010). Therefore, families should not be considered as isolated entities; they should also include meaningful and influential ties with friends, kin, in-laws, neighbors, colleagues, and others (Bengtson, 2001; Widmer et al., 2013). Thus, family ties are bound neither to the nuclear family unit nor to the limits of the household. They are of high relevance, exerting both positive and negative effects on various domains of life (Widmer, 2010). Conjugal conflicts, for instance, are significantly associated with network characteristics in which the conflicted parties are embedded (Widmer et al., 2004). Related research has shown that the likelihood of conflicts is associated with the composition of one's personal network. Networks that focus on the nuclear family tend to produce more conflicts than do more diverse networks (Widmer et al., 2018).

This research hypothesizes that family interventions trigger complex dynamics in conflict patterns within personal networks involving ties that go beyond the boundaries of the nuclear family unit. A dyadic view on conflicts in family intervention research does not account for the diversity of conflict patterns that are present in families receiving child welfare services.

The importance of trusting working relationships for conflict resolution

Although conflict resolution is a common goal of family intervention programs, research reveals inconsistent results regarding the effects on interpersonal conflicts. Whereas some studies reveal a significant decrease in conflicts between family members (Lewis, 2005; Messmer et al., 2021; Smith, 1995), others show only a negligible decrease (Al et al., 2014) or no effects at all (van Puyenbroeck et al., 2009). Child protection service and family intervention practitioners describe their work with families afflicted by intense conflicts as particularly demanding, as they are affected by higher levels of stress, intense emotions, and feelings of helplessness and frustration (Katz et al., 2019; Saini et al., 2012; Sudland, 2020). Working with high-conflict families requires more time, energy, and resources than does working with other cases (Saini et al., 2012). Parents and children who are involved in or affected by intense parental conflicts are often in difficult emotional states and experience great distress (Katz et al., 2019). Some social workers have reported parents pressuring their children to act or decide on their behalf (Saini et al., 2019).

In social work, a good relationship between practitioners and service users is of paramount importance for achieving the intervention goals (Ruch et al., 2010). Extensive research suggests that service users’ motivations and outcomes of social work interventions are largely determined by the quality of the practitioner–client relationship, which is also referred to as working alliance (Al et al., 2014; Buckley et al., 2011). However, in child welfare services, establishing a positive relationship is often challenging due the mandatory nature of the intervention (Ferguson et al., 2021) and the power asymmetries in the practitioner–client relationship (Dumbrill, 2006).

Among the several aspects that define a relationship, mutual trust is essential to build a working alliance between practitioners and service users (Bekaert et al., 2021; Cossar et al., 2016; Husby et al., 2018; Valenti et al., 2020). Research concerning the practitioner–parent relationships points out that a trusting relationship is associated with increased cooperation (Messmer et al., 2021; Thrana & Fauske, 2014). Regarding the practitioner–child relationship, trust is an essential requirement for the participation of children in family interventions (Cossar et al., 2016; Husby et al., 2018). In general, research has shown that a trusting relationship between family members and practitioners is associated with higher commitment of families and more positive outcomes (Bekaert et al., 2021).

As a matter of fact, research also indicates that the beneficial effects of a trusting working relationship on intervention outcomes particularly come into play for the resolution of family conflicts (Sudland, 2020). The development of trust in the context of family conflicts is a fundamental requirement for positive change. A recent study has shown that practitioners who take sides in mother–child conflicts fail in releasing the tension because both parties cannot trust the practitioner (Brauchli, 2021, p. 358). Therefore, this study expects that the effects on interpersonal conflicts can be better explained by considering the level of mutual trust between the practitioner and the multitude of individuals involved in a particular family intervention rather than by an isolated view on dyadic relationships, such as the practitioner–parent or practitioner–child relationship.

A network approach to evaluating family interventions

This research proposes a network approach to examine the effects of family interventions on interpersonal conflicts. When applying such an approach to measure the effects of family interventions, one must consider two main elements: first, the personal network in which parents are embedded, as its structure and composition can greatly influence parenting resources, beliefs, and practices (Belsky, 1984); second, the intervention, which consists of the ties between the practitioner and the members of the network who are involved in the family intervention. The goal of the practitioner is to support the network to self-sufficiently develop new strategies to secure and promote the well-being of the child(ren). In such a framework, the practitioner is deemed a “relational resource” that can be consulted by the network members (Folgheraiter, 2004). In this regard, research suggests that the establishment of a trusting working relationship results in effective participation of network members in the family intervention (Cossar et al., 2016; Husby et al., 2018; Valenti et al., 2020).

Personal network

Personal networks of parents are diverse in terms of size, composition, and structure (Widmer, 2006, 2007). They consist of meaningful and interconnected relationships that go beyond the boundaries of the nuclear family. These relationships provide a crucial socialization context for children since they are affected by them in many positive and negative ways (Cochran & Brassard, 1979). A network approach stresses that family problems are rooted in personal relationships because they are the consequences of shared decisions and actions. Therefore, effective solutions to such problems must emerge from networks as a joint, rather than an individual effort.

A network approach offers a variety of meaningful indices to investigate interpersonal conflict, which is the central theme of the present study. First, conflict density refers to the number of interactions within the network that are characterized as conflict-ridden. Research has shown that high conflict density has negative effects on the psychological health of the network members (Widmer et al., 2018). However, family relationships seldom consist of conflicts exclusively; rather, they are related to support relationships, which is expressed in ambivalence (Aeby et al., 2020). Ambivalence refers to the ongoing negotiation of obligations, expectations, and autonomy within family ties, which is an inherent part of how families organize themselves (Connidis, 2015). Second, conflict reciprocity indicates that a negative behavior toward a network member is actively reciprocated with a negative response. High conflict reciprocity refers to networks where conflicts are intense. Conflict reciprocity usually occurs in more intimate relationships. In marital relationships, for instance, conflict reciprocity is associated with aggression, dissatisfaction, and distress (Ramos Salazar, 2015). However, intense conflicts may also reflect the effort to maintain relationships in times of acute stress, which prevents groups from dissolution (Coser, 1956). Third, centrality is a measure that indicates to what extent an individual plays a central role in conflicts. High centrality within conflict relationships suggests that an individual is in active opposition to significant network members (Widmer et al., 2018) and is considered a common enemy within that network (Everett & Borgatti, 2014). Individuals with a high centrality likely play a critical part in the diffusion of negative tensions and feelings in networks (Bowen, 1993). Research has shown that high centrality of individuals in family conflicts is associated with stress and poorer psychological health (Widmer et al., 2018).

Family intervention

In family interventions, practitioners establish working relationships with several individuals who are part of the intervention (e.g., parents, children, and grandparents). Each relationship between the practitioner and a family member is unique but also mutually dependent. For example, social workers who work with family members dealing with strong disputes describe their intermediary position as highly challenging due to the risk of getting involved in a loyalty conflict (Sudland, 2020). From a relational standpoint, family interventions are complex and demanding because practitioners must create trusting working relationships with several family members who are connected through ambivalent ties. However, the relationship between the practitioner and the family members who are involved in the intervention is the main mechanism for conflict resolution. According to Bowen’s (1993) family systems theory, the tensions that arise from a conflict between two individuals can be relieved via the principle of “triangulation”. When two family members are in conflict, the practitioner can reduce the tension by establishing a trusting relationship with both parties. By doing so, the practitioner transforms the dyadic relationship of the two conflicted family members into a triad, thereby relieving the stress. In families that are enrolled in interventions, numerous dyadic conflicts are often present. Therefore, practitioners are a part of various interlocking triads. Additionally, a trusting relationship with network members is important because it promotes the development of a shared attitude toward the family intervention within the larger social context of a family. The involvement of network members stimulates the communication in the network in favor of the goals and actions of the intervention (Perry et al., 2016).

Hypotheses

In light of the proposed network approach, it is expected that a trusting relationship between the practitioner and the individuals involved in the family intervention will have a decreasing effect on three manifestations of interpersonal conflicts in the personal network of the parents. A trusting working relationship is expected to (1) decrease the number of interactions in the personal network of the parent that are associated with conflicts (conflict density), (2) lower the number of intense conflicts (conflict reciprocity), and (3) decrease the centrality of parents in interpersonal conflicts (betweenness centrality of the parent).

Data and measurement

Data collection and participants

Data for the present study were collected in two waves from the German speaking part of Switzerland using paper-and-pencil questionnaires. Access to the families was facilitated by 18 service providers of Social Pedagogical Family Support—the most frequently used home-based service in the Swiss child welfare system (Wetzel et al., 2020). This service includes parent counseling, joint reflections and discussions with the family, and activities with children. Common goals of this type of intervention are the development of parenting skills, promotion of children's competencies in various domains, and resolution of parent–child conflicts (Messmer et al. 2021). The service providers were instructed to invite every family that was referred to their service between June 2018 and April 2019 to participate in the study. The data collection included a questionnaire for the parents and another one for the practitioners (hereafter called social pedagogues). The parents provided information about their personal network, and the social pedagogues about the family intervention. Each party filled in a questionnaire at the start of the intervention and a second one about 6 months later. The social pedagogues who were performing the home visits collected the data for the first and second wave. On average the questionnaires for wave one were filled in during the fourth home visit (M = 4.12, SD = 1.8). All parents who agreed to participate signed an informed consent form that outlined the responsibilities of each party in the study. In total, 70 parents were interviewed in the first wave. However, 24 dropped out of the study before the start of the second wave, 13 dropped out because the intervention was terminated, 6 refused to participate in the second wave, and 5 did not complete the second questionnaire. No significant differences were found concerning the network size and conflict variables between the parents who dropped out and the final sample, which consisted of 46 parents.

The average duration between the first and the second assessment was 198.4 days (SD = 32.3), during which the social pedagogues visited the families on average 29.3 times (SD = 14.0) for 114.4 min (SD = 35.4) per home visit. Slightly more than one-third of the parents (39.1%) were mandatorily referred to the family intervention, while the others agreed to it. Fathers and mothers comprised 13.0% and 87.0% of the sample, respectively, of which 47.8% were single parents. The average age of the parents at the start of the intervention was 35.2 years (SD = 8.0). The number of children per family was 2.2 (SD = 0.9) with an average age of 9.2 years (SD = 5.6). More than half of the parents were Swiss citizens (56.5%); 54.3% was unemployed and 60.9% received social welfare benefits. Furthermore, 37% of the parents did not have vocational education and 15.2% only had minimal German language skills. Additionally, almost one-third of all parents (32.6%) reported that they had been diagnosed with a mental illness. The socioeconomic and health-related characteristics reveal that the parents in this sample were affected by multiple disadvantages.

Measurement

Information about the parents’ personal networks was collected using the Personal Network Method (PNM). This validated instrument allows the collection of ego-centric network data (Widmer et al., 2013). Respondents are asked to list a maximum of 14 individuals who are important to them based on the following question: Who are the individuals who, over the past year, have been very important to you, even if you have not gotten along well with them? The respondents then provided information about not only their relationships to the listed individuals (e.g., relationship type, frequency of contact) but also the individuals themselves (e.g., age, sex). Lastly, respondents were asked to provide information about the conflictual relationships in their personal network considering ego-alter and alter-alter relationships as follows: Conflicts and tensions appear in every relationship from time to time; who upsets whom frequently? This data allows the computation of structural network measures such as density, reciprocity, or centrality and to draw graphical representations of personal networks (Crossley et al., 2015).

The family intervention was documented with a second paper-and-pencil questionnaire that was completed by the social pedagogue who conducted the intervention. During the second assessment, the social pedagogues not only examined the parents’ list of personal network members and indicated the ones they involved since the beginning of the intervention but also indicated the frequency of contact and level of mutual trust separately for each involved network member.

Dependent variables

The effects of the family intervention on interpersonal conflicts were analyzed using three different sets of personal network members. The first set, the respondents’ in-neighborhood, contained only the network members who directly upset the respondent. The second set, the out-neighborhood, included only network members who were directly upset by the respondent. Finally, the third set, the full network, considered all members of the respondents’ personal network. Within these network sets, the following measures were computed and transformed to change scores (t1–t0):

Conflict density: The conflict density is the number of actual ties in a network divided by the total possible number of its ties. Hence, the density measure has a value between 0 and 1. For the in-neighborhood subset, the density measure considered only the number of possible ties between the network members who directly upset the respondent (including the respondent); and vice versa for the out-neighborhood subset, with higher conflict density indicating a denser pattern of interactions between network members that are associated with conflict.

Conflict reciprocity: Conflict reciprocity measures the fraction of reciprocated ties over all actual ties, which is technically known as “arc reciprocity”. The value ranges between 0 and 1, with higher values indicating the presence of more conflict reciprocity in the network.

Betweenness centrality: The betweenness centrality counts how often a node lies on the shortest path between two nodes. Its standardized measure ranges between 0 and 1, with higher values indicating a more central position in a set of interpersonal conflicts.

Independent variables

Intervention effects on these conflict measures were analyzed using multiple linear regression models. The models contain three independent variables: (1) the level of mutual trust, (2) the frequency of contact between the social pedagogue and the respondents’ network members, (3) the percentage of the respondents’ network members involved in the intervention—and the baseline score of the dependent variable as a control variable. The independent variables consider only the ties between the social pedagogue and the respondent's network members, without including the working relationship with the respondent. Therefore, the intervening variables can be interpreted more clearly as the effects of involving the respondent's personal network.

The level of mutual trust between the social pedagogue and the network members was rated by the former during the second assessment using the following question: How strong do you rate the mutual trust between you and the network member? The social pedagogue rated the level of trust for every network member who was involved separately on a 5-point scale (very low, low, mediocre, strong, and very strong). For the regression analysis, the mean score of all network members involved in an intervention was calculated.

Frequency of contact between the social pedagogue and the network members was also measured separately for every network member involved, rated on a 7-point scale (once during the past 6 months, several times during the past 6 months, once a month, several times a month, once a week, several times a week, and daily). Frequency of contact covered face-to-face as well as telephone- or internet-based contact (How frequently did you have contact with the network member?). Analogous to the level of mutual trust, the mean score of all network members per case was used for the regression models.

Network involvement was measured by calculating the percentage of members of the respondents’ personal network who were included in the intervention by the social pedagogue between its start and the second assessment.

Finally, the baseline score of the dependent variable was included as an independent variable to control for the regression toward the mean effect (Allison, 1990). This effect describes the phenomena of extremely low or high values at a given time being closer to the average of the sample at the following measurement point. Thus, the baseline value in the regression models is expected to be negatively associated with the dependent variables. It is important to stress that the effects of the baseline value should not be interpreted as intervention effects.

Graphical representation of personal networks and the intervention

Figures 1 and 2 present a visualization of the independent and dependent variables included in the regression models, respectively. Figure 1 shows the personal network of a mother (focal individual) with a daughter (aged 6) and a son (aged 11) at the start of the intervention (t0) and roughly 6 months later (t1). The edges in the network represent conflicts between its members. The personal network at t0 consists of nine members including the mother, her two children, two sisters, two friends, ex-husband, and a neighbor. Five out of these nine network members are involved in conflicts. The arrows imply the direction of the conflicts. For instance, the focal individual frequently upsets both of her children and her ex-husband, while her son and her ex-husband frequently upset the focal individual. At the second assessment (t1), changes in the network size, composition, and structure can be detected. The network now consists of seven members instead of nine, with the focal individual listing both of her children, her neighbor, and one of her sisters who were all present in the network at t0. However, the ex-husband, the two friends, and one sister are no longer listed. Instead, her new partner and the mother of the new partner are now part of the personal network. Regarding the structural changes, the daughter is not involved in conflicts anymore. The new partner and his mother, neither of whom were not in the network at t0, now upset the son frequently. The neighbor, who was not involved in conflicts at the first assessment, now regularly upsets the new partner of the focal individual.

Figure 1.

Graphical representation of interpersonal conflicts at the start of the intervention (t0) and 6 months later (t1).

Figure 2.

Graphical representation of frequency of contact, level of mutual trust, and network involvement.

Figure 2 shows that, in this case, the social pedagogue involved the daughter, son, and partner of the focal individual in the intervention. Hence, the neighbor, the sister, and the mother of the new partner were not involved. Regarding the frequency of contact with the network members, this figure indicates that the social pedagogue had slightly more contact with the daughter than with the son and the new partner of the focal individual during the first 6 months of the intervention, during which the highest level of mutual trust was developed between the social pedagogue and the daughter, followed by the partner and the son. Overall, the frequency of contact as well as the level of mutual trust with personal network members is rather low, when considering the measurement scales.

Results

Table 1 shows the mean scores and the SDs of the dependent variables used in the regression models at the start of the intervention and the size of each network set. The in-neighborhood subset, comprising all individuals who directly upset the respondent, primarily includes the respondent's children, parents, siblings, partner, and ex-partner(s). The most frequently named network members who are directly upset by the respondent are their children, parents, siblings, and partner. The entire network also consists predominantly of ties that were mentioned in the first two subsets, with the addition of friends, in-laws, and kin. According to Table 1, conflict density is substantially higher when only individuals who directly upset the respondent are included and vice versa. Not only conflict density but also respondents’ centrality in interactions associated with conflict is higher in the in- and out-neighborhood set in comparison to the entire network. However, the conflict reciprocity measure shows that the intensity of conflicts is almost equal in all the network sets, indicating that roughly half of the conflict relationships are reciprocal.

Table 1.

Descriptive statistics of conflict measures at the start of the intervention.

	Alters who upset the focal individual		Alters who are upset by the focal individual		All network members of the focal individual
	M	SD	M	SD	M	SD
Size of the network set	3.70	2.13	3.20	2.28	8.98	3.38
Conflict density	0.54	0.32	0.50	0.40	0.21	0.22
Conflict betweenness centrality	0.20	0.30	0.16	0.25	0.07	0.13
Conflict reciprocity	0.52	0.45	0.54	0.45	0.55	0.36

Table 2 shows the frequency of contact and the level of mutual trust between the social pedagogue and the network members between the start of the intervention and the second assessment. When considering all 46 cases included in the sample, the social pedagogues involved a total of 144 network members (not counting the respondents). Thus, 3.12 (SD = 1.42) network members on average, which is 46.7% (SD = 26.1%) of the network members, were involved. The level of mutual trust was unrateable in 10 cases due to the young age of some children or other reasons (n = 134). The results in Table 2 are categorized by the network members who were most involved: the current partner, ex-partner, children, and parents. The category “others” contains all other network members who did not fit in one of the existing categories. The social pedagogues had most frequent contact with the respondents’ children (M = 4.16). They also rated the level of mutual trust with children the highest (M = 3.46) compared to the other groups. The contact frequency with the current partner or the ex-partner is similar (M = 2.95/3.14), but the mutual trust was rated higher between the social pedagogue and the current partner (3.32) than with the ex-partner (M = 2.86). The lowest contact frequency is between the social pedagogues and the respondents’ parents and with the heterogeneous group “others” (M = 2.18). However, the level of mutual trust is higher in the latter case (M = 3.36).

Table 2.

Relationship attributes between the social pedagogues and network members at t1.

	Frequency of contact between the social pedagogue and network members of the focal individual (7-point scale)		Level of mutual trust between the social pedagogue and network members of the focal individual (5-point scale)
	M	SD	M	SD
Partner (n = 19/19)	2.95	1.31	3.32	0.75
Children (n = 90/84)	4.16	1.27	3.46	0.84
Ex-partner (n = 7/7)	3.14	1.21	2.86	0.90
Parents (n = 11/10)	2.18	1.47	2.80	0.79
Others (n = 17/14)	2.18	1.38	3.36	0.84
Total (n = 144/134)	3.56	1.52	3.35	0.84

Spearman correlations were performed between the baseline scores of the conflict measures and the intervention variables (Table 3). Spearman's rho coefficient indicated a moderate positive correlation between the frequency of contact and the level of mutual trust between the social pedagogues and the network members (r_s = 0.33, p = .023, n= 46) and a negative association between the level of trust and the baseline score of out-neighborhood betweenness centrality (r_s = −0.32, p = .028, n = 46). Due to their moderate strength, these two significant correlations do not violate the rule of multicollinearity. In line with other research (Sapin et al., 2016), numerous conflict measures are positively intercorrelated within and between the three network sets; because they are used in separate regression models, they do not violate the rule of multicollinearity.

Table 3.

Spearman correlations of baseline scores of the conflict measures and intervention variables.

	1	2	3	4	5	6	7	8	9	10	11
IN density	1
IN betweenness centrality	.170	1
IN reciprocity	.784***	.626***	1
ON density	.724***	.244	.675***	1
ON betweenness centrality	.221	.896***	.628***	.104	1
ON reciprocity	.647***	.491***	.766***	.891***	.406**	1
FN density	.542***	.293*	.599***	.507***	.333*	.537***	1
FN betweenness centrality	.163	.884***	.560***	.224	.834***	.439**	.452**	1
FN reciprocity	.576***	.401**	.695***	.591***	.415**	.688***	.392**	.362*	1
INT frequency of contact	.163	.066	.161	.114	.071	.023	.204	.113	.031	1
INT level of mutual trust	.127	-.264	-.070	.144	-.324*	.001	.191	-.240	-.083	.334*	1
INT network involvement	−.065	−.033	−.104	−.135	.016	−.180	.131	.058	−.197	.083	.237

Note. FN = full network; IN = in-neighborhood, INT = intervention; ON = out-neighborhood.

* p ≤ .05, **p ≤ .01, ***p ≤ .001.

To analyze the effects of the family intervention on the interpersonal conflicts in the respondents’ personal networks, a series of linear multiple regression models were tested (Table 4). The level of mutual trust shows the expected effect of lowering conflicts in personal networks across various measures. Higher levels of mutual trust with network members significantly decrease conflict density and conflict reciprocity in all three network sets. No effects regarding the centrality of the focal individual in interpersonal conflicts were observed.

Table 4.

Multiple linear regression models predicting changes in conflict measures between t0 and t1.

	Alters who upset the focal individual		Alters who are upset by the focal individual		All network members of the focal individual
	B	SE	B	SE	B	SE
Conflict density (DV)
Frequency of contact	0.154***	0.040	0.100*	0.048	0.054*	0.021
Level of mutual trust	−0.243**	0.080	−0.213*	0.097	−0.087*	0.041
Percentage of network members involved	0.159	0.171	0.418	0.207	0.183*	0.088
Baseline score of the dependent variable	−0.707***	0.140	−0.904***	0.135	−0.543***	0.112
Conflict betweenness centrality (DV)
Frequency of contact	−0.001	0.032	−0.020	0.027	0.021	0.013
Level of mutual trust	−0.071	0.066	−0.054	0.058	−0.042	0.026
Percentage of network members involved	0.043	0.135	−0.002	0.116	−0.009	0.055
Baseline score of the dependent variable	−0.920***	0.121	−0.935***	0.127	−0.878***	0.109
Conflict reciprocity (DV)
Frequency of contact	0.146**	0.051	0.085	0.055	0.038	0.042
Level of mutual trust	−0.279**	0.101	−0.281*	0.111	−0.196*	0.083
Percentage of network members involved	0.343	0.215	0.421	0.242	0.170	0.181
Baseline score of the dependent variable	−0.802***	0.127	−0.950***	0.139	−0.695***	0.131

Note. DV = dependent variable.

*p ≤ .05, **p ≤ .01, ***p ≤ .001.

Contrary to the level of mutual trust, the frequency of contact with network members has the opposite effect on interpersonal conflicts. High contact frequency is associated with increased conflict density in all three network sets. However, significant effects on conflict reciprocity can only be reported in the in-neighborhood model.

Of less importance is the percentage of network members involved in the family intervention. Network involvement increases conflict density, but only in the entire network. Therefore, involving more network members increases the likelihood of negative ties between alters in the personal network of the focal individual during the intervention.

Discussion

This research hypothesized that a trusting working relationship between practitioners and the personal network members of parents in a family intervention is associated with a decrease in various conflict measures. The results presented in Table 4 generally confirm this hypothesis while also providing further insights into how interpersonal conflicts are impacted by a family intervention.

Mutual trust between practitioners and network members is associated with a significant decrease in conflict density and conflict reciprocity. Hence, network members are less affected by internal conflicts in the medium term when a trusting relationship is developed with the social pedagogue. Lower conflict density implies less negative valences in personal networks, which are associated with a less stressful environment for the network members (Widmer et al., 2018). In this regard, research suggests that lower conflict density decreases the likelihood of emotional insecurity of children about family relationships and developmental problems (Cummings et al., 2015).

The results related to conflict reciprocity complement these findings. A trusting working relationship between practitioners and network members has a calming effect on intense conflicts. One reason for this could be that the members of the personal network developed better communication skills, which is a common goal in this type of family intervention. The lack of such skills is a common cause for conflict reciprocity (Ramos Salazar, 2015). Ethnographic observations of home visits have shown that practitioners who simultaneously work with several family members are more successful in stimulating collective and more self-sufficient communication about conflicts (Messmer et al. 2021). However, these effects could also be attributed to the presence of the practitioner, who may have become a significant member of the respondent's personal network during the intervention. This would indicate that the network members were not able to deal with interpersonal conflicts themselves and so preferred to integrate the practitioner in their conflict-solving practices. In this regard, qualitative research would help provide an in-depth understanding of the relationship between the role of the practitioner and the patterns of interpersonal conflicts in families. Studies with postintervention measures report different findings regarding the sustainability of conflict indicators. A study by Lewis (2005) showed that positive achievements of an intensive short-term intervention were maintained 8 to 10 months after the intervention. The study reported that family members were screaming less at each other and experienced less anger. Furthermore, parents preserved their improved skills of handling children when they yell or swear. In contrast, a study that also investigated a short-term family intervention reported no significant changes in parent–child conflicts after its completion (Al et al., 2014).

No effects of mutual trust were detected on the centrality of respondents’ in interpersonal conflicts. This indicates that a trusting working relationship with network members affects the conflict dynamics in personal networks but does not affect the structural position of the focal individual. In this regard, research has shown that many practitioners put the parents at the centre of social pedagogical family interventions (Messmer et al., 2021). This could have the effect of parents remaining in the position of control, having to manage conflicts and tensions. Higher levels of centrality indicate that parents are simultaneously involved in conflicts with several individuals or groups. Being in such a position in conflict networks not only makes parenting more demanding and stressful because of the need to moderate several conflicts but also makes parents a target of hostility and criticism. Research has shown that individuals who play a central role in family conflicts are more likely to be affected by higher levels of distress (Sapin et al., 2016).

In contrast to the positive effects of mutual trust, a high frequency of contact with network members did not release tensions in personal networks of parents. This result is in line with other research in the field of family interventions. A study by Littell (1997) revealed that a higher intensity of family preservation services was associated with an increase in the recurrence of child maltreatment and higher rates of out-of-home placements. According to the author, this could be related to the fact that intensive interventions were especially provided to high-risk families, which anyway had only a small likelihood of success. As research has shown, home visits can be perceived as intrusive by family members and can cause distress (Cook, 2020). Having frequent contact with network members could trigger several conflicts simultaneously, resulting in higher conflict density within the network due to the clashing of interests of several individuals. As a matter of fact, frequency of contact is not a crucial factor for the establishment of mutual trust. High congruence regarding the intervention goals and expectations between family members and the practitioner, the practitioner's emphatic communication style, and a working relationship that is on an equal footing are some of the more important factors (Rüegger et al., 2021).

In comparison to mutual trust and frequency of contact, the percentage of network members involved in the family intervention barely affects interpersonal conflicts. This result suggests that resolving a specific conflict within a network may have a positive effect on conflicts between other network members. This indicates that the positive effects of mutual trust might be diffused within the personal network and stop further chains of conflicts (de Bel et al., 2021).

Limitations of the study

The present research has several limitations that should be discussed. One major limitation is the measurement to capture the level of mutual trust, which only represents the view of the social pedagogue. As the term “mutual” implies, a trusting relationship involves the social pedagogue's and the service user's perspective on this matter. Therefore, it is possible that the network members involved in the family intervention may have rated the level of trust differently. The second limitation concerns the exclusive focus on conflicts. Research has shown that the occurrence of solely negative ties in family networks is rather rare. Personal networks consist of positive and negative ties, resulting in ambivalent relational patterns (Widmer, 2010). Research suggests that ambivalent relationships cause more distress than do exclusively conflicted ties (Uchino et al., 2004). Ambivalence emerges in a variety of patterns associated with different levels of psychological distress (Aeby et al., 2020; Sapin et al., 2016). Therefore, future studies should consider both positive and negative ties in their research designs. The third limitation of this study is the sample size, which is rather small (n = 46). A small sample size increases the likelihood of not accounting for small effects (type II error). Finally, it would be beneficial to include postintervention data to examine whether changes in interpersonal conflicts are due to the presence of the social pedagogue or whether they are sustainable.

Conclusions and practical implications

This study contributes to family intervention research by adopting a unique network approach, which offers novel findings about intervention effects. The proposed approach stresses the importance of considering the diversity of personal ties and their larger social contexts in family interventions. The results reveal that the level of mutual trust between practitioners and the individuals involved in a particular family intervention differs considerably. This highlights the uniqueness of each relationship when working with several network members. The results suggest that practitioners must establish mutual trust with not only parents but also other individuals to promote a less stressful family environment. A network approach takes the role of the practitioner in family interventions into account. In this regard, it encourages practitioners to promote conflict resolution capacities of families instead of becoming a permanent and indispensable part of conflict resolution practices. The characteristics of the study sample reveal that many families are affected by larger structural inequalities. Because interpersonal conflicts are often associated with socioeconomic disadvantages, practitioners should also support families in gaining access to services that address pressing socioeconomic needs.

Footnotes

Ethics

Ethical approval was not required for this research. However, the study participants signed an informed consent form with detailed information about the study and the rights and responsibilities of each party involved.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Swiss National Science Foundation (grant number 172598).

Acknowledgements

The author would like to thank Professor Eric D. Widmer for his valuable feedback on the manuscript.

ORCID iD

Lukas Fellmann

References

Aeby

Gauthier

J-A

Widmer

E. D.

(2020). Patterns of support and conflict relationships in personal networks and perceived stress. Current Sociology, 69(7), 1–21. https://doi.org/10.1177/0011392120948926

C. M. W.

Stams

G. J. J. M.

Asscher

J. J.

van der Laan

P. H.

(2014). A programme evaluation of the Family Crisis Intervention Program (FCIP): Relating programme characteristics to change. Child & Family Social Work, 19(2), 225–236. https://doi.org/10.1111/j.1365-2206.2012.00896.x

Allison

P. D.

(1990). Change Scores as Dependent Variables in Regression Analysis. Sociological Methodology, 20, 93–114. https://doi.org/10.2307/271083

Autorengruppe Kinder- und Jugendhilfestatistik (2021). Kinder- und Jugendhilfereport Extra 2021: Eine kennzahlenbasierte Kurzanalyse. Dortmund. DJI/TU Dortmund. https://doi.org/10.3224/84742240

Bekaert

Paavilainen

Scheke

Baldacchino

Jouet

Zablocka-Zytka

Bachi

Bartoli

Carra

Cioni

Crocamo

Appleton

(2021). Family members’ perspectives of child protection services, a metasynthesis of the literature. Children and Youth Services Review, 128, 1–13, Article No. 106094, https://doi.org/10.1016/j.childyouth.2021.106094

Belsky

(1984). The determinants of parenting: A process model. Child Development, 55(1), 83–96. https://doi.org/10.2307/1129836

Bengtson

V. L.

(2001). Beyond the nuclear family: The increasing importance of multigenerational bonds. Journal of Marriage and the Family, 63(1), 1–16. https://doi.org/10.1111/j.1741-3737.2001.00001.x

Bowen

(1993). Family Therapy in Clinical Practice. Jason Aronson Inc.

Brauchli

(2021). Das Wohl der Kinder und die Selbstbestimmung der Eltern: Eine qualitative Untersuchung zur Sozialpädagogischen Familienbegleitung in der Schweiz. Beltz Juventa.

10.

Buckley

Carr

Whelan

(2011). ‘Like walking on eggshells’: Service user views and expectations of the child protection system. Child & Family Social Work, 16(1), 101–110. https://doi.org/10.1111/j.1365-2206.2010.00718.x

11.

Cochran

Brassard

J. A.

(1979). Child development and personal social networks. Child Development, 50(3), 601–616. https://doi.org/10.2307/1128926

12.

Connidis

I. A.

(2015). Exploring ambivalence in family ties: Progress and prospects. Journal of Marriage and Family, 77(1), 77–95. https://doi.org/10.1111/jomf.12150

13.

Cook

L. L.

(2020). The home visit in child protection social work: Emotion as resource and risk for professional judgement and practice. Child & Family Social Work, 25(1), 18–26. https://doi.org/10.1111/cfs.12647

14.

Coser

L. A.

(1956). The Functions of Social Conflict. Race, class and social structure. Routledge.

15.

Cossar

Brandon

Jordan

(2016). ‘You've got to trust her and she's got to trust you’: Children's views on participation in the child protection system. Child & Family Social Work, 21(1), 103–112. https://doi.org/10.1111/cfs.12115

16.

Crossley

Bellotti

Edwards

Everett

M. G.

Koskinen

Tranmer

(2015). Social network analysis for ego-nets. Sage.

17.

Cummings

E. M.

Koss

K. J.

Davies

P. T.

(2015). Prospective relations between family conflict and adolescent maladjustment: Security in the family system as a mediating process. Journal of Abnormal Child Psychology, 43(3), 503–515. https://doi.org/10.1007/s10802-014-9926-1

18.

de Bel

Snijders

T. A. B.

Widmer

E. D.

(2021). Modelling ambivalent triads in family research. Social Work Research, 98, 1–15. Article No. 102577. https://doi.org/10.1016/j.ssresearch.2021.102577

19.

Dumbrill

G. C.

(2006). Parental experience of child protection intervention: A qualitative study. Child Abuse & Neglect, 30(1), 27–37. https://doi.org/10.1016/j.chiabu.2005.08.012

20.

Everett

M. G.

Borgatti

S. P.

(2014). Networks containing negative ties. Social Networks, 38, 111–120. https://doi.org/10.1016/j.socnet.2014.03.005

21.

Ferguson

Disney

Warwick

Leigh

Cooner

T. S.

Beddoe

(2021). Hostile relationships in social work practice: Anxiety, hate and conflict in long-term work with involuntary service users. Journal of Social Work Practice, 35(1), 19–37. https://doi.org/10.1080/02650533.2020.1834371

22.

Folgheraiter

(2004). Relational social work: Toward networking and societal practices. Jessica Kingsley Publishers.

23.

Grych

J. H.

Fincham

F. D.

(Eds). (2001). Interparental conflict and child development: Theory, research, and applications. Cambridge University Press. https://doi.org/10.1017/CBO978051 1527838

24.

Husby

I. S. D.

Slettebø

Juul

(2018). Partnerships with children in child welfare: The importance of trust and pedagogical support. Child & Family Social Work, 23(3), 443–450. https://doi.org/10.1111/cfs.12435

25.

Katz

Glucklich

Piller

Matty

D. E.

(2019). Between chaos and danger: Spotlighting social workers’ views in cases of child maltreatment accusations in the context of high-intensity parental disputes. Children and Youth Services Review, 107, 1–7. Article No. 104497. https://doi.org/10.1016/j.childyouth.2019.104497

26.

Krishnakumar

Buehler

(2000). Interparental conflict and parenting behaviors: A meta-analytic review. Family Relations, 49(1), 25–44. https://doi.org/10.1111/j.1741-3729.2000.00025.x

27.

Lewis

R. E.

(2005). The effectiveness of families first services: An experimental study. Children and Youth Services Review, 27(5), 499–509. https://doi.org/10.1016/j.childyouth.2004.10.009

28.

Littell

J. H.

(1997). Effects of the duration, intensity, and breadth of family preservation services: A new analysis of data from the Illinois family first experiment. Children and Youth Services Review, 19(1-2), 17–39. https://doi.org/10.1016/S0190-7409(97)00004-2

29.

Messmer

Wetzel

Fellmann

Käch

(2021). Sozialpädagogische Familienbegleitung: Ausgangsbedingungen – Praxis – Wirkungen. Beltz Juventa.

30.

Perry

B. L.

Pullen

Pescosolido

B. A.

(2016). At the intersection of lay and professional social networks: How community ties shape perceptions of mental health treatment providers. Global Mental Health, 3, 1–17. Article No. e3. https://doi.org/10.1017/gmh.2015.25

31.

Ramos Salazar

(2015). The negative reciprocity process in marital relationships: A literature review. Aggression and Violent Behavior, 24, 113–119. https://doi.org/10.1016/j.avb.2015.05.008

32.

Ruch

Turney

Ward

(2010). Relationship-based social work: Getting to the heart of practice. Jessica Kingsley Publishers.

33.

Rüegger

Gautschi

Becker-Lenz

Rotzetter

(2021). Bedeutung und Aufbau von Vertrauen in der Sozialpädagogischen Familienbegleitung. Gesellschaft – Individuum – Sozialisation. Zeitschrift Für Sozialisationsforschung, 2(2), 1–16. https://doi.org/10.26043/GISo.2021.2.3

34.

Saini

Black

Godbout

Deljavan

(2019). Feeling the pressure to take sides: A survey of child protection workers’ experiences about responding to allegations of child maltreatment within the context of child custody disputes. Children and Youth Services Review, 96, 127–133. https://doi.org/10.1016/j.childyouth.2018.11.044

35.

Saini

Black

Lwin

Marshall

Fallon

Goodman

(2012). Child protection workers’ experiences of working with high-conflict separating families. Children and Youth Services Review, 34(7), 1309–1316. https://doi.org/10.1016/j.childyouth.2012.03.005

36.

Sapin

Widmer

E. D.

Iglesias

(2016). From support to overload: Patterns of positive and negative family relationships of adults with mental illness over time. Social Networks, 47, 59–72. https://doi.org/10.1016/j.socnet.2016.04.002

37.

Sarrazin

Cyr

(2007). Parental conflicts and their damaging effects on children. Journal of Divorce & Remarriage, 47(1-2), 77–93. https://doi.org/10.1300/J087v47n01_05

38.

Smith

M. K.

(1995). Utilization-Focused evaluation of a family preservation program. Families in Society: The Journal of Contemporary Social Services, 76(1), 11–19. https://doi.org/10.1177/104438949507600102

39.

Sudland

(2020). Challenges and dilemmas working with high-conflict families in child protection casework. Child & Family Social Work, 25(2), 248–255. https://doi.org/10.1111/cfs.12680

40.

Thrana

H. M.

Fauske

(2014). The emotional encounter with child welfare services: The importance of incorporating the emotional perspective in parents’ encounters with child welfare workers. European Journal of Social Work, 17(2), 221–236. https://doi.org/10.1080/13691457.2013.798628

41.

Uchino

B. N.

Holt-Lunstad

Smith

T. W.

Bloor

(2004). Heterogeneity in social networks: A comparison of different models linking relationships to psychological outcomes. Journal of Social and Clinical Psychology, 23(2), 123–139. https://doi.org/10.1521/jscp.23.2.123.31014

42.

Valenti

Celedonia

K. L.

Wall-Parker

Strickler

(2020). Trust is essential: Identifying trust building techniques from youth providers across the service array. Children and Youth Services Review, 117, 1–8. Article No. 105295. https://doi.org/10.1016/j.childyouth.2020.105295

43.

van Eldik

W. M.

Haan

A. D. de

Parry

L. Q.

Davies

P. T.

Luijk

M. P. C. M.

Arends

L. R.

Prinzie

(2020). The interparental relationship: Meta-analytic associations with children's maladjustment and responses to interparental conflict. Psychological Bulletin, 146(7), 553–594. https://doi.org/10.1037/bul0000233

44.

van Puyenbroeck

Loots

Grietens

Jacquet

Vanderfaeillie

Escudero

(2009). Intensive family preservation services in Flanders: An outcome study. Child & Family Social Work, 14(2), 222–232. https://doi.org/10.1111/j.1365-2206.2009.00626.x

45.

Wetzel

Messmer

Fellmann

(2020). Hilfen zur Erziehung. In Bonvin

J.-M.

Hugentobler

Knöpfel

Maeder

Tecklenburg

(Eds), Wörterbuch der Schweizer Sozialpolitik (pp. 235–237). Seismo Verlag.

46.

Widmer

E. D.

(2006). Who are my family members? Bridging and binding social capital in family configurations. Journal of Social and Personal Relationships, 23(6), 979–998. https://doi.org/10.1177/0265407506070482

47.

Widmer

E. D.

(2007). Social capital in wide family contexts: An empirical assessment using social network methods. International Review of Sociology, 17(2), 225–238. https://doi.org/10.1080/03906700701356861

48.

Widmer

E. D.

(2010). Family configurations: A structural approach to family diversity. Ashgate.

49.

Widmer

E. D.

Aeby

Sapin

(2013). Collecting family network data. International Review of Sociology, 23(1), 27–46. https://doi.org/10.1080/03906701.2013.771049

50.

Widmer

E. D.

Girardin

Ludwig

(2018). Conflict structures in family networks of older adults and their relationship with health-related quality of life. Journal of Family Issues, 39(6), 1573–1597. https://doi.org/10.1177/0192513X17714507

51.

Widmer

E. D.

Kellerhals

Levy

(2004). Types of conjugal networks, conjugal conflict and conjugal quality. European Sociological Review, 20(1), 63–77. https://doi.org/10.1093/esr/20.1.63

Effects of family interventions on interpersonal conflicts: A network perspective

Abstract

Summary

Findings

Applications

Keywords

Introduction

The importance of trusting working relationships for conflict resolution

A network approach to evaluating family interventions

Personal network

Family intervention

Hypotheses

Data and measurement

Data collection and participants

Measurement

Dependent variables

Independent variables

Graphical representation of personal networks and the intervention

Results

Discussion

Limitations of the study

Conclusions and practical implications

Footnotes

Ethics

Funding

Acknowledgements

ORCID iD

References