Next Article in Journal
Perceptions and Responses of Adolescents and Youth in the Spanish Protection System to Bullying and Gender-Based Bullying Situations, and Their Relationship with the Training Received
Previous Article in Journal
Loneliness and Social Withdrawal Among College Students: The Mediating Role of Internet Addiction and the Moderating Effect of Sex
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Adolescents
Volume 5
Issue 4
10.3390/adolescents5040052
adolescents-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Presentation and Initial Validation of a New Observational Situation and Coding System for Assessing Triadic Family Interactions with Adolescents

by
Michaël Romet
1,
Nicolas Favez
1,2,
Amalia Foletta
1,
Annie Burnier
1,
Aleksandra Mrozek
1,
Marie Schumacher
1 and
Hervé Tissot
1,2,*
1
Center for Family Studies, University Institute of Psychotherapy, Department of Psychiatry, Lausanne University Hospital and University of Lausanne, Lausanne 1011, Switzerland
2
Faculty of Psychology and Educational Sciences, University of Geneva, Geneva 1205, Switzerland
*
Author to whom correspondence should be addressed.
Adolescents 2025, 5(4), 52; https://doi.org/10.3390/adolescents5040052
Submission received: 3 July 2025 / Revised: 21 August 2025 / Accepted: 19 September 2025 / Published: 25 September 2025
(This article belongs to the Section Adolescent Health and Mental Health)
Browse Figure
Versions Notes

Abstract

This study investigated the validity of a new observational procedure for assessing mother–father–adolescent conflict interactions. Addressing key gaps in the field, this procedure moves beyond mother–adolescent dyads and self-report questionnaires by offering a standardized, observational family-level approach. The procedure combines the Lausanne Trilogue Play—Conflict Discussion Task (LTP–CDT) with a coding system, the Family Conflict and Alliance Assessment Scales with Adolescents (FCAAS). In a sample of 82 two-parent families with adolescents aged 10 to 13, the FCAAS demonstrated excellent inter-rater reliability and good validity. Confirmatory factor analysis supported a theoretically grounded two-factor structure. Criterion and construct validity were confirmed via self-reports of marital satisfaction and coparenting, respectively. Ecological validity was supported by family members’ self-reports. This tool offers promising research and clinical applications and may be extended to diverse populations to assess known-group validity.

1. Introduction

In daily family life, a teenager interacts with their parents in various dyadic and triadic relational configurations: with the mother only, with the father only, with one parent in the presence of the other, with both parents at the same time, or simply witnessing a discussion between the parents. More complex configurations are also possible if the child has siblings. Among these relational configurations, researchers first focused on the mother–child dyad, before broadening their attention to the father–child dyad, mutual influences between family dyads, and family-level processes such as coparenting (i.e., the coordination and teamwork between two adults regarding childrearing) [1]. Family-level processes—interactions involving both parents and at least one child—have been shown to represent a level of analysis in their own right in the study of family functioning, and to predict unique proportions of variance in children’s and adolescents’ psychological outcomes beyond parenting-related variables [2,3,4]. In particular, coparenting characterized by low cooperation and a high degree of conflict between parents has been linked to externalizing problems in adolescents, especially in early adolescence [5]. Moreover, as they involve both parents, family-level processes have been described as being closely related to (but yet distinct from) the quality of the marital relationship. For example, several studies reported statistical associations between coparenting and marital satisfaction [6].
Observation of families “in action” has shown to be a useful method for assessing the quality of family relationships [7]. When looking at the existing observational tools to assess family-level processes during family interactions, coding systems focus on various aspects of family functioning, such as family competence or cohesion [8,9]. Nevertheless, these tools entail two main issues. First, they are not designed for the assessment of many of the relational configurations that can be found in daily family life, such as when a parent and a teen discuss in the presence of the other parent. However, including these configurations might inform researchers about the organization and hierarchy in family systems. Second, the tasks and coding systems generally focus on coparenting in families with younger children [10] and do not specifically allow the assessment of coparenting in families with adolescents. This latter point is critical, since it has been found to be relevant to characterize family-level relationships and predict various developmental outcomes in adolescents [11].
The family alliance (FA) model provides a theoretical framework that has been developed to study triadic family interactions with infants, including all possible relational configurations and the coparenting relationship. FA can be defined as the coordination between family members when they perform a joint activity such as playing, sharing a meal, or discussing [12,13]. FA assessment is based on a macroanalysis of several verbal and non-verbal cues, such as postures, gaze orientation, interferences, turn-taking, adaptiveness of child stimulation, and mutual smiles [14]. Historically, triadic interactions have been systematically evaluated in the laboratory with the Lausanne Trilogue Play (LTP), a standardized observational situation where two parents are asked to play with their infant following a four-part scenario, with each part reflecting one of the four relational configurations that are possible within a triad: First, one parent is asked to play with the child, while the other parent is asked to remain “simply present”; parents switch roles; all three family members play together; and finally, parents discuss while the child is left on her own for a little while. The assessment distinguishes three types of FA: cooperative, conflicted, and disordered. Cooperative FA refers to family interactions in which all family members coordinate and share positive affects. Conflicted FA refers to family interactions in which a coparenting or marital conflict overtly or covertly prevents the triad from sharing a genuine moment of quality. Disordered FA refers to family interactions where (self-) exclusion prevails, which typically prevents the triad from coordinating and sharing positive affects. Most importantly, the latter two types of FA have been found to predict increased rates of psycho-functional disorders in infants, toddlers, and preschoolers and are thus considered as “problematic” [15,16]. In a longitudinal study, the evolution pattern of FA in early childhood was also found to predict the level of social cognition skills of 15-year-old adolescents [3].
Although the FA model has been developed for early family interactions, it has also been used in an exploratory way for assessing families with adolescents in both clinical and research contexts. The LTP situation has been used as a diagnostic and therapeutic tool in family therapy, and links between the quality of family interactions during the LTP situation have been associated with adolescent psychopathology [17]. Nonetheless, despite numerous preliminary studies using the LTP situation for families with adolescents, FA assessment has yet to be properly validated in the context of families with adolescents.
For the assessment of FA in adolescence, we decided to use a conflict discussion task [18] instead of a cooperative task for the assessment of FA in infancy and childhood. Indeed, family conflict might serve as a suitable process for evaluating the observed quality of family relationships. On the one hand, it is a central family-level process in adolescence [19] and a normative mechanism of change as adolescents grow toward autonomy [20]. On the other hand, when conflicts are intense and frequent or when conflict management is steeped in hostility (i.e., destructive instead of constructive conflict) [21], it might be linked to potentially harmful effects on adolescent adjustment and health [22]. This rationale led us to design the Lausanne Trilogue Play—Conflict Discussion Task (LTP–CDT), an observational situation that aims to elicit conflict interactions and negotiation in the mother–father–adolescent triad.
To assess interactions occurring during the LTP–CDT, we developed the Family Conflict and Alliance Assessment Scales—with adolescents (FCAAS) [23]. The FCAAS is a macro-analytical observational coding system based on the Family Alliance Assessment Scales (FAAS) [14], the validated instrument used to assess the quality of FA in infancy. However, we needed to adapt the FAAS for obvious reasons. First, family interactions with adolescents may differ from those with infants in terms of the child’s contribution to the interaction, since adolescents may be more verbally active (i.e., they can use language), which can lead them to interact in ways other than infants, including being defensive, critical, or sarcastic. Second, as both tasks differ (play vs. discussion), they will elicit different behaviors. For example, assessing cooperation during a play might be based on coordination of attention or actions, whereas assessing coordination during a discussion might be based on verbal cues. In addition to these adaptations linked to the targeted age and the nature of the task, the literature in the fields of adolescent development and parent–adolescent communication was screened to find constructs and scales that could be relevant for the assessment of the quality of family interactions during a triadic conflict discussion task. Several FCAAS scales were inspired by this literature (for a complete list of references, see Supplementary Material S1) and adapted to meet the complexity of triadic family interaction patterns.
In the present study, we introduce the Lausanne Trilogue Play—Conflict Discussion Task (LTP–CDT) and the Family Conflict and Alliance Assessment Scales—with adolescents (FCAAS). We conducted this study in a sample of families with early adolescents, as research suggests that links between family-level processes and adolescent outcomes are stronger during early adolescence than in later stages [5]. The aims of this study were to examine the reliability of the FCAAS scale, assess its construct and criterion validity, and evaluate the ecological validity of the LTP-CDT. Interrater reliability analyses were conducted to test reliability. Criterion validity was examined through associations between FCAAS scores and parents’ marital satisfaction, while construct validity was assessed through associations between FCAAS scores and parents’ self-reported quality of coparenting. Ecological validity of the LTP-CDT was evaluated based on mothers’, fathers’, and adolescents’ reports of how typical the interactions during the task were compared to those in their daily family life.

2. Materials and Methods

2.1. Sample Characteristics

A sample of N = 87 non-referred mother–father–adolescent triads was recruited in 2022 from the general population in the area of Lausanne, Switzerland. Flyers were posted in the medical offices of pediatricians and general practitioners, and on the hospital’s website. The study was conducted in accordance with the Declaration of Helsinki and with the Swiss Federal Human Research Ordinance (HRO), and approved by the Cantonal Commission on Ethics in Human Research of the State of Vaud (CER-VD; Project n. 2021-01859, 25 October 2021). Inclusion criteria were as follows: families had to have adolescents aged between 10 and 13 years and to be fluent in French (the language of the questionnaires). Families were included only if none of the members had a known psychopathological diagnosis. Five families out of the initial 87 were excluded because of missing observational data. The final sample of adolescents was n = 33 girls and n = 49 boys with a mean age of 12.04 years (SD = 0.94). Mothers’ ages ranged from 34 to 60 years with a mean age of 44.04 years (SD = 4.21). Fathers’ ages ranged from 33 to 59 years with a mean age of 45.99 years (SD = 4.97). None of the families were in the lower socioeconomic class, 3.7% were in the lower-middle class, 9.8% were in the middle class, 17.1% were in the upper-middle class, and 69.5% were in the upper class. Parents were separated in four families and divorced in one family (this triad was measured with the stepfather). The adolescent was a single child in 12.2% of cases, and there were 30.5% of youngest, 11.0% of middle, and 46.3% of eldest children. All family interactions were in French, except for two in English and one in Spanish. These three families were fluent in French and could complete the questionnaires; however, they were asked to discuss in their usual language so that interactions would be as typical as possible.

2.2. Procedure

Families visited the laboratory after having given their written consent. Family interactions were filmed during the Lausanne Trilogue Play—Conflict Discussion Task (LTP–CDT). Directly after the interaction, each family member was given a paper-and-pencil questionnaire to assess their perception of the typicality of the interaction. In the following two weeks, each parent and the adolescents were asked to complete online questionnaires using the REDCap electronic data capture tools [24].
The LTP–CDT procedure is a standardized observational situation that can occur in the laboratory or at home. The three family members are seated on chairs, each positioned approximately 120 cm from the other two chairs (measured from the center of one chair to the center of the other), to create an equilateral interactive triangle. This arrangement serves (1) to place family members relatively close to each other and encourage dialog, and (2) to ensure that they are equidistant from each other.
The instructions occur in three parts. First, instructions are given to select the “hottest” topic of conflict within the family. The experimenter mentions several potential subjects and explains that the family can select a conflict occurring between the teenager and one parent, between the teenager and both parents, between the parents about the teenager’s education, or between all three family members. Second, the rest of the instructions are given regarding the four-part scenario of the LTP–CDT (i.e., one parent discusses with the adolescent while the other parent remains simply present, then parents switch roles, then parents discuss together while adolescent remains simply present, and all three discuss together). Finally, family members are given the opportunity to ask any questions they might have. Once the family is ready, the experimenter leaves the room and gives an audio signal from the control room to indicate the start of the family interaction as well as the transitions to the next part of the LTP–CDT scenario. Each part of this scenario lasts for exactly 3 min. Detailed instructions given to LTP–CDT’s participants can be found in Supplementary File S2.
In this validation study, interactions were filmed using two cameras positioned opposite each other. The order of the first two LTP–CDT parts was randomized and counterbalanced, meaning that half of the families started with the mother–adolescent (+father) configuration, while the remaining families started with the father–adolescent (+mother) configuration. Regarding the selected conflictual subjects, families in our study discussed the following: screen time and use (n = 24), school and homework (n = 13), relationship with siblings and parental management of conflicts between siblings (mostly parental unfairness; n = 11), compliance with the rules and setting of new rules (n = 10), parent–adolescent relationship regarding communication and conflict (n = 9), storage and cleaning at home (n = 7), and other (n = 9).

2.3. Measures

2.3.1. Family Conflict and Alliance Assessment Scales—With Adolescents (FCAAS)

The FCAAS comprises 10 scales: Postures & gazes, turn-taking, mutual respect for coparenting roles, conflict resolution, affective climate, mentalization, role reversal, coparenting support, autonomy promotion, and adolescent autonomy. They are summarized in Table 1, and a more thorough description can be found in Supplementary File S1.
The scales were initially developed by the first author, drawing on a literature review of adolescent development and family relationships, as well as on the FA model originally formulated for families with young children [12]. They were subsequently refined and calibrated through the review of several videotaped parent–adolescent conflict discussions. The manual was then thoroughly reviewed by the second and last authors until a consensus was reached among them.
The first author used the FCAAS to assess the quality of family interactions in the complete sample at the end of data collection. To evaluate inter-rater reliability, a second coder (third author) was trained for 8 h by the first author. For feasibility reasons, approximately one third of the sample was selected for double-coding. An initial reliability check was conducted after the first set of ten families. At that point, as reliability was satisfactory for all scales except Postures and gazes, the first and second authors engaged in an additional 8 h of training with a specific focus on this scale. The initial scores for the first ten families were nevertheless retained for the final analyses. The second coder subsequently coded eighteen families to reach a total of n = 28 double-coded families (34% of the sample). The reliability scores improved slightly between the first and second set of families, although the inter-rater reliability still remained slightly lower than expected for Postures and gazes (see Section 3.2 for more details). At the end of the coding process, the scores from the two coders were averaged to resolve any disagreements.

2.3.2. Relationship Assessment Scale

The Relationship Assessment Scale (RAS) [25] is a 7-item questionnaire that assesses the degree of marital satisfaction in a romantic relationship. It has been adapted and validated in French [26]. It includes questions such as “How well does your partner meet your needs?” or “In general, how satisfied are you with your relationship?” Each item is rated on a 5-point Likert scale with scores ranging from 1 (low satisfaction) to 5 (high satisfaction). A total score is obtained by computing the sum of the items, with a higher score indicating higher marital satisfaction. Both parents filled in the RAS separately (Cronbach’s α = 0.81 for mothers and α = 0.75 for fathers). A positive correlation was expected, with higher scores of marital satisfaction being linked with higher scores on the FCAAS instrument.

2.3.3. Coparenting Inventory for Parents and Adolescents

The Coparenting Inventory for Parents and Adolescents (CIPA) [27,28] is a self-report questionnaire that assesses coparenting based on 25 items ranging along three dimensions (cooperation, conflict, and triangulation), each measured with three different targets (mothers’ and fathers’ contribution to coparenting, as well as coparenting in the parental dyad). Agreement with the affirmations ranged from 0 (not at all true) to 4 (completely true), and a mean score was computed for each dimension. High scores corresponded to a high presence of the given phenomenon, i.e., high cooperation or high conflict. To maintain parsimony in the models, we decided to only perform analyses on the target of the parental dyad (leaving aside the specific contribution of each parent). Each dimension at the dyadic level consisted of 4 items, with a total of 12 items for the three dimensions (mothers: Cronbach’s α = 0.84 for cooperation, α = 0.68 for conflict, α = 0.74 for triangulation; fathers: α = 0.63 for cooperation, α = 0.67 for conflict, α = 0.66 for triangulation). Each parent answered the questions separately. We expected the FCAAS instrument to correlate positively with coparenting cooperation, and negatively with coparenting conflict and triangulation.

2.3.4. Typicality Questionnaire

To assess the ecological validity of family interactions during the LTP–CDT, we operationalized ecological validity in terms of typicality, that is, the extent to which each member of the triad assesses the situation and the behaviors of all partners as representative of everyday life [29]. We adapted the questionnaire from Favez et al. [29] for use with the LTP–CDT and we developed a comparable but shorter adolescent version. The questionnaire was completed by each family member after the end of the LTP–CDT. For parents, the questionnaire comprised 13 five-point Likert items, with answers ranging from 1 (=not [typical] at all) to 5 (=completely[typical]). The first item focused on parents’ general perception of typicality compared with daily interactions in the family. The remaining 12 items focused on the extent to which each family member behaved in a “normal” way during each of the four parts of the LTP–CDT scenario. For adolescents, the questionnaire comprised four five-point Likert items similar to those of the parents, except that the normality of behaviors was assessed only for the interaction as a whole and not regarding each part of the LTP–CDT scenario. An average typicality score was computed for each family member. We expected that interactions would be representative of everyday interactions [29].

2.4. Statistical Analyses

First, we computed a set of descriptive statistics and bivariate correlations for all study variables. Second, inter-rater reliability was investigated according to the recommendations of Ten-Hove and colleagues [29] to estimate intraclass correlation coefficients (ICCs) in a way that allows for variance due to differing numbers of raters per subject, (i.e., incomplete or unbalanced designs). Using the absolute agreement between raters to test for inter-rater reliability, we calculated the ICCs (A, khat) [30] and variance terms with maximum likelihood estimation (MLE). Here and for all analyses, average ratings were used for double-coded family interactions and single ratings of the first author were used for the remaining family interactions. Third, we used confirmatory factor analysis (CFA) to assess the factor structure of the scale within a structural equation modelling (SEM) framework. On the basis of on the unidimensionality of the FA model in infancy [14], we started with the assumption that a one-factor would likely be a sound factor structure. Prior to conducting the CFA, we evaluated the adequacy of our sample size using a Monte Carlo simulation in Mplus (1000 replications, N = 82) based on the hypothesized one-factor model. The simulation examined the distribution of the χ2 test of model fit across replications, comparing observed and expected percentiles. Results indicated that the empirical Type I error rate was close to the nominal value (5%), and fit indices were well-approximated, suggesting sufficient power to detect model misfit given the model’s degrees of freedom. Complete simulation output is provided in the Supplemental Material (Document S3). After this simulation study, we ran the analysis to estimate this one-factor model. Because the results of this model (see Section 3.3) were unsatisfactory, we used modification indices (MI) to refine our model and test an alternative model with two factors (FA and coparenting). Fourth, we assessed the links between the FCAAS and a set of control variables using the two-factor structure of the FCAAS. The selected control variables were age and gender of the adolescent, family socioeconomic status, family situation (parents together vs. not together), the conflictual subject chosen for the LTP–CDT, and the scenario of the LTP–CDT (which parent starts in the first part of the LTP–CDT). Because the conflictual subject chosen for the LTP–CDT is a nominal variable, we tested whether there were differences regarding the FCAAS scores with one-way ANOVAs (one for each factor of the FCAAS). Fifth, we assessed the construct validity of the FCAAS by investigating the links between fathers’ and mothers’ marital satisfaction (RAS scores) and the two-factor structure of the FCAAS. Mothers’ and fathers’ RAS scores were added as covariates in the two-factor CFA model of the FCAAS, with freely estimated covariance paths between the FCAAS factors and the RAS scores, as well as between both parents’ RAS scores. Then, similar to the fifth step, we used the two-factor model and investigated the links between fathers’ and mothers’ perceptions of coparenting (CIPA scores) and the factor structure of the FCAAS. The six CIPA scores were added as covariates in the FCAAS two-factor CFA model. The model was specified with freely estimated covariances between all these variables, including the covariances between all included variables of the CIPA. Finally, we assessed the reported typicality of family interactions during the LTP–CDT. Mean scores were computed for the perceptions of each family member, and internal consistency was assessed using Cronbach’s alpha.
Descriptive statistics and bivariate correlations were performed with IBM SPSS 29, whereas inter-rater reliability indices were calculated in R using the scripts made available by Ten-Hove and colleagues [30] on the Open Science Framework (https://osf.io/8j26u/, accessed on 5 May 2023). Mplus (version 8.9, Muthén & Muthén, Los Angeles, CA, USA) was used for confirmatory factor analyses [31]. To assess the fit of the SEM models, we used the chi-square test, confirmatory fit index (CFI), Tucker–Lewis index (TLI), root mean square error of approximation (RMSEA), and standardized root mean squared residual (SRMR). Indicators for good-fitting models are a non-significant chi-square, a CFI and TLI above 0.95 for excellent fit (at least above 0.90 for acceptable fit), an RMSEA below 0.06 (at least below 0.08 for acceptable fit), and an SRMR below 0.08 [32]. Modification indices are considered significant above the 3.84 threshold and when indices of the fully standardized expected parameter change (stdYX EPC) are above the cutoff criterion of 0.20 [33]. We handled the missing data with pairwise correction.

3. Results

3.1. Descriptive Statistics and Bivariate Correlations

Descriptive statistics and bivariate correlations between all study variables are presented in Supplementary Table S1.

3.2. Inter-Rater Reliability

The results of inter-rater reliability are presented in Table 2. Reliability was excellent for the scales “Conflict resolution” and “Role reversal” (ICCs = 0.90), good for the scales “Turn-taking”, “Mutual respect for coparenting roles”, “Affective climate”, “Mentalization”, “Coparenting support”, “Autonomy promotion”, and “Adolescent autonomy” (ICCs > 0.81), and lower but still in the moderately acceptable range for the scale “Postures & gazes” (ICC = 0.64).

3.3. Factor Structure Validity

The Monte Carlo simulation (1000 replications, N = 82) for the hypothesized one-factor model indicated that the empirical Type I error rate for the χ2 test closely approximated the nominal 0.05 level (observed proportion = 0.090, expected = 0.050). The expected χ2 value at the 0.05 percentile was 49.802, while the observed value was 53.588, indicating only a slight upward deviation. Across the distribution, expected and observed percentiles were similar, supporting adequate statistical power and model stability at this sample size. Moreover, the fit indices averaged over the 1000 replications were RMSEA = 0.029 (with an RMSEA ≤ 0.05 found in 72% of the replications), CFI = 0.983 (with a CFI ≥ 0.95 found in 95% of the replications), and SRMR = 0.042 (with an SRMR ≤ 0.05 found in 92% of the replications). These results suggested that our sample size was adequate to conduct the intended analyses. The complete simulation output is provided in the Supplemental Materials.
The results of the CFA for the one-factor model indicated only a moderate adjustment of the model to the data: Chi-square (χ2) = 79.866, degrees of freedom (df) = 35, p < 0.001, CFI = 0.873, TLI = 0.837, SRMR = 0.083, RMSEA = 0.125, 90% CI [0.089, 0.161]. MI requested on this one-factor model indicated that there could be a significant fit improvement if covariances were added on the residuals of the scales “Mutual respect for coparenting roles” and “Role reversal” (MI = 16.799; fully standardized expected parameter change (stdYX EPC) = 0.486), as well as on the residuals of the scales “Role reversal” and “Coparenting support” (MI = 13.233; stdYX EPC = 0.450). These results suggested that the three scales “Mutual respect for coparenting roles”, “Role reversal”, and “Coparenting support”, might constitute a separate factor, which made sense from a theoretical point of view, since these three scales all relate to the concept of coparenting. Stemming from these results, we ran a two-factor model with one FA factor (using the following scales as indicators: “Postures & gazes”, “Turn-taking”, “Conflict resolution”, “Affective climate”, “Mentalization”, “Autonomy promotion”, and “Adolescent autonomy”), and one coparenting factor (using the following scales as indicators: “Mutual Respect for coparenting roles”, “Role reversal”, and “coparenting support”). This model showed a better adjustment to the data: χ2 = 54.806, df = 34, p = 0.013, CFI = 0.941, TLI = 0.922, SRMR = 0.078, RMSEA = 0.086, 90% CI [0.040, 0.127]. The CFI and TLI values indicated an adequate (albeit not excellent) fit, while the SRMR value indicated a good fit. In contrast, the chi-square test was significant, and the RMSEA value was slightly above the cutoff for acceptable fit. This last result might be due to our small sample size [31,33]. Considering its theoretical relevance and its adequate fit on average, the two-factor model was deemed satisfactory. As shown in Figure 1, the factor loadings ranged from 0.353 to 0.860 for the FA factor and from 0.480 to 0.886 for the coparenting factor, with all standardized estimates being significant (all p-values < 0.002). The estimation of the correlation between the two factors was high, r = 0.73, p < 0.001). Finally, the Cronbach’s alphas for the items composing the FA and coparenting dimensions (α = 0.87 and α = 0.75, respectively) indicated good internal consistency.
As the “Postures and gazes” scale showed both a weak interrater reliability and low (although significant) factor loading in the two-factor CFA model, we decided to conduct additional analysis to estimate the one- and two-factor models on nine scales, removing the “Postures and gazes” scale. As the models were non nested, we referred to the ΔCFI, ΔRMSEA, and ΔSRMR to evaluate the differences between the models. Reference cutoffs selected to determine the best models were a ΔCFI, a ΔRMSEA, and a ΔSRMR ≥ 0.010 [34,35], Results of the estimations of these models showed that most differences between the models with 9 and 10 items fell below these cutoffs, except for two: For the one-factor model, the value of the ΔRMSEA was 0.011 in favor of the model using ten items. For the two-factor model, the value of the ΔCFI was 0.010 in favor of the model using nine items. As these results did not offer strong arguments in favor of removing the “Postures and gazes” scale, we decided to retain this scale. Moreover, this scale was essential for theoretical reasons as allowed us to bridge the gap between the FA model [12] used as a theoretical basis of this study and the assessment model we present in this study.

3.4. Control Variables

The results of the estimation of the model showed a poor fit (χ2 = 111.483, df = 74, p = 0.003, CFI = 0.903, TLI = 0.876, SRMR = 0.084, RMSEA = 0.079, 90% CI [0.046, 0.107]), as only the CFI and RMSEA indices were acceptable. Such a poor fit makes sense given that the correlations of all control variables with the FA and the coparenting factors ranged from −0.077 to 0.214 and were not significant, except for the correlation regarding the FA factor and adolescent gender (r = 0.27, p = 0.011; higher means on the FA factor in families with boys) and the correlation regarding the coparenting factor and family situation (r = −0.21, p = 0.027; higher means on the coparenting factor in families in which parents are together compared to the one in which parents are separated or divorced). Regarding the conflictual subject discussed during the LTP-CDT, scores at the FA factor significantly differed, F(6,75) = 2.61, p = 0.023, and so did the scores at the coparenting factor, F(6,75) = 2.91, p = 0.012. Post hoc tests (Bonferroni) showed that FA scores were significantly lower (p = 0.010) when families discussed the parent–adolescent relationship (M = 19.39, SD = 5.36) than when they discussed the relationship with siblings and parental management of conflicts between siblings (M = 28.91, SD = 4.39). In addition, coparenting scores were significantly lower when families discussed the parent-adolescent relationship (M = 8.83, SD = 3.02) than when they discussed the sibling relationship (M = 12.73, SD = 1.69), p = 0.049, or the subject of school and homework (M = 13.04, SD = 2.15), p = 0.015.

3.5. Criterion Validity: Links Marital Satisfaction

The results of the estimation of the model showed an acceptable fit (χ2 = 75.659, df = 50, p = 0.011, CFI = 0.933, TLI = 0.912, SRMR = 0.071, RMSEA = 0.078, 90% CI [0.038, 0.112]). Although the chi-square test was significant, the CFI, TLI, SRMR, and RMSEA values were still in the acceptable range. In this model, the FA factor significantly correlated with fathers’ RAS scores (r = 0.34, p = 0.002), whereas it did not significantly correlate with mothers’ RAS scores (r = 0.18, p = 0.103). The coparenting factor correlated significantly with fathers’ RAS scores (r = 0.33, p = 0.003), and with mothers’ RAS scores (r = 0.36, p = 0.015). Mothers’ and fathers’ marital satisfaction correlated positively and significantly (r = 0.31, p = 0.009).

3.6. Construct Validity: Links Coparenting Perceived by the Parents

Estimation of the model revealed an acceptable fit (χ2 = 117.099, df = 82, p = 0.007, CFI = 0.939, TLI = 0.911, SRMR = 0.065, RMSEA = 0.070, 90% CI [0.038, 0.098]). Although the chi-square test was significant, the CFI, TLI, SRMR, and RMSEA values were acceptable. In this model, the FA factor significantly correlated with fathers’ perceptions of coparenting cooperation (r = 0.39, p = 0.001), conflict (r = −0.29, p = 0.010), and triangulation (r = −0.34, p = 0.005), and with mothers’ perceptions of coparenting cooperation (r = 0.34, p = 0.001), conflict (r = −0.34, p = 0.002), and triangulation (r = −0.27, p = 0.036). The coparenting factor significantly correlated with perceptions of coparenting cooperation by fathers (r = 0.49, p < 0.001) and mothers (r = 0.40, p = 0.002). It also correlated significantly with fathers’ and mothers’ perceptions of coparenting conflict (r = −0.28, p = 0.027; r = −0.28, p = 0.019, respectively) and triangulation (r = −0.36, p = 0.009; r = −0.47, p < 0.001, respectively).

3.7. Ecological Validity

To assess ecological validity, we examined the mean scores on the Typicality scales completed by all the participants, with potential total scores ranging from 1 (=not [typical] at all) to 5 (=completely[typical]). In this sample, the scores ranged between 1 and 5, with a relatively high mean for all participants (mothers: M = 3.92, SD = 0.76; fathers: M = 4.00, SD = 0.71; adolescents: M = 3.92, SD = 0.78). These results suggested that parents and adolescents found, on average, that the interactions they experienced in the LTP-CDT were quite typical as compared to similar kind of interactions at home.

4. Discussion

This paper presents a new procedure that combines an observational situation, the LTP–CDT, and its coding system, the FCAAS. This procedure extends the FA model from early childhood to adolescence and allows researchers in the field of family psychology to assess the quality of interactions in mother–father–adolescent triads. The description of the assessment and rating procedures was followed by a study that provided convincing evidence for the reliability and validity of the FCAAS, as well as for the ecological validity of the LTP–CDT.
First, this validation study achieved inter-rater reliability for the rating scales, which was good to excellent for nine of ten rating scales. This result supports the high reliability of the FCAAS instrument, especially since we used a strict criterion (absolute agreement between raters) to assess it. Regarding the scale “Postures & Gazes”, the reliability was in the moderate range, which indicates that clarification of criteria and specification of coding rules might be required. Future studies may indicate whether this scale is meaningful and potentially more reliable with additional training. Should the opposite occur, it might be eliminated from the coding system to improve its precision and time consumption.
Second, the results of the CFA showed that the factor structure of the FCAAS seemed to comprise two factors referred to as “Family Alliance” and “coparenting”. This result was surprising at first, given that theory and available data on family triads with infants have pointed to the one-dimensionality of FA- and coparenting-related behavioral indicators in the FAAS [14]. One explanation for this differentiation could be that coparenting-related and family processes are independent because they exist at two different hierarchical levels, with the family pertaining to the system level and coparenting to the subsystem level [36]. Another explanation for the differentiation between FA and coparenting might be that an adolescent child is much more autonomous and active than a baby in social interaction. Therefore, the parental dyad might progressively differentiate itself from whole-family functioning as the family transitions from childhood to adolescence. Indeed, family functioning might shift from a “mother–father–child” configuration in early childhood to a “mother–father + adolescent” configuration in adolescence.
Third, total scores on the FA and coparenting dimensions were not related to adolescent age, family socioeconomic status, or the scenario of the LTP–CDT, whereas they were related to adolescent gender, family situation, or the conflictual subject of the LTP–CDT. Regarding adolescent gender, there was a surprising effect according to which families with boys scored higher on the FA factor. We imagine that such an effect might stem from differences in parenting and coparenting according to the child’s gender [11], and therefore, future research should most certainly include this control variable in analyses. Regarding family situation, scores at the coparenting factor were lower in families in which the parents were divorced or separated, which seems logical given the links between marital and coparenting functioning [37]. Regarding the subject that was selected for the LTP–CDT, families discussing the parent–adolescent relationship displayed lower FA or coparenting scores than when discussing other subjects. Such differences can be explained by the fact that some conflictual subjects may be more concrete or easier to solve than others. In addition, the family’s subject selection (i.e., the type, difficulty, or intensity of conflict) might thus be indicative of the quality of family relationships.
Fourth, the two-factor structure of the FCAAS showed adequate criterion validity, as several significant associations between the FCAAS factors and marital satisfaction appeared. For the coparenting factor of the FCAAS, better observed functioning of the coparenting relationship was related to higher marital satisfaction for both parents, which is in line with the literature [37]. For the FA factor of the FCAAS, better-quality family interactions were related to fathers’ higher marital satisfaction, whereas there were no significant associations with mothers’ marital satisfaction. This gender difference might suggest that while marital satisfaction and coparenting might be inseparable for both fathers and mothers, marital satisfaction and the quality of family relationships might be more disconnected in mothers than in fathers. Moreover, gender differences in marital satisfaction might be absent or small in terms of quantity (i.e., level of satisfaction) but exist in terms of quality. Indeed, husbands and wives might differ regarding their reasons behind marital (in)satisfaction or regarding the links with other variables within or outside the couple and the family [38].
Fifth, the two-factor structure of the FCAAS showed adequate construct validity, which was tested using a self-report measure of coparenting. Indeed, FA and observed coparenting scores correlated significantly with all scores of fathers’ and mothers’ perceptions of coparenting. All correlations between FA or observed coparenting scores and reported coparenting went in the expected direction, meaning that better scores on our observational rating scales were related to better scores in coparenting cooperation, or lower scores in coparenting conflict or triangulation. These results show that the FCAAS instrument has good construct validity, and that lower scores on the instrument might indicate actual difficulties in the coparenting relationship.
Sixth, participants reported that family interactions performed during the LTP-CDT were highly typical, which supports the ecological validity of this new observational situation. The LTP-CDT seems to be prone to generate behaviors in the laboratory that are representative of daily behaviors in the family. This is in line with the fact that patterns of communication in families with adolescents have been repeated for many years, which makes interactional patterns, behaviors, and attitudes very instinctive and hard to control. Additionally, the fact that the conflict to solve was chosen by the family probably added to the impression of the family members that they were in a real-life discussion. Therefore, our results indicate that family interactions during the LTP–CDT may not be sensitive to social desirability.
There were several limitations to this study. The first limitation was the small sample size although it was still acceptable for this type of research. We aim to compensate in the future with further investigations on new samples of families. The second limitation is the high homogeneity of our sample, given that two-thirds of the families were in the upper-middle or upper socioeconomic class and less than 15% of the families belonged to lower classes. The third limitation of our validation study is that our sample was limited to preadolescents (aged between 10 and 13). However, the age of adolescents in our sample was not linked to the scores on the FCAAS scales (see analyses of the control variables), which suggests that the use of the LTP–CDT and the FCAAS might be extended to adolescents older than 13 years. Future studies should confirm our results for the entire adolescent developmental period. The fourth limitation was that only one child per family was included, even in households with multiple children. The complexity of family interactions increases exponentially when additional participants, such as siblings, are considered. Our study focused on the first degree of family complexity, involving both parents and a target child. It is also important to acknowledge that each child in a family may experience coparenting differently, and there is no single coparenting model that applies uniformly across siblings [39]. Consequently, the observed links between family-level processes and adolescent outcomes may not fully capture the dynamics experienced by other children in the household. The last limitation is that the coefficients of internal reliability for some CIPA scores were at the lowest values in the acceptability range, i.e., slightly below 0.70. Therefore, although they showed strong associations with other study variables that were consistent with what could be expected, results including these CIPA scores should be interpreted with caution. Based on these few limitations, the next stage in this line of research would be to test the LTP–CDT and FCAAS in larger samples and with populations different from the sample in this study, such as families with older adolescents, low-income families, families with a referred member, and families with multiple children. Most importantly, future research should investigate the extent to which FCAAS scores predict adolescents’ psychological outcomes. Such studies would benefit from including both referred and non-referred adolescents, thereby enabling a rigorous test of the scale’s known-groups validity. In addition, the LTP–CDT and FCAAS could be tested in families with same-sex parents so that our assessment tool may be considered more generally, as targeting parent–parent–adolescent interactions (instead of only mother–father–adolescent interactions).
Finally, a topic of future research might be the use of the LTP–CDT and FCAAS in the clinical setting. Potentially, our task and coding system could provide clinicians with a standardized way to assess communication problems that could be worked upon to improve the quality of family relationships and ultimately the adolescent prognosis. For instance, the LTP–CDT could be used at the beginning and end of family therapy and thus provide a measure of the evolution of the family. Importantly, the use of this instrument with referred families may require several adaptations. For instance, the clinician might want to moderate the instruction concerning the selection of the “hottest” topic for the LTP–CDT but might instead suggest a recurring topic based on prior therapeutic sessions with the family. We would nevertheless strongly recommend against the use of the LTP–CDT with families where there is a risk of violence. Also, another potential use of the LTP–CDT and FCAAS might be to use it as an intervention tool by providing video-feedback to the family, i.e., watching the video-recorded family interaction with the family and discussing meaningful excerpts with the aim of supporting the family by emphasizing their multiple resources [40]. Finally, although the use of this research tool might be difficult to implement in day-to-day clinical practice (e.g., time consumption), the benefits may outweigh the costs. The next steps might include a dialog with clinicians and empirical investigations regarding the utility and effectiveness of using the LTP–CDT for clinical purposes.
In conclusion, the present study introduced a new observational tool that could help researchers to further investigate the quality of mother–father–adolescent interactions. The evidence presented here suggests that the designed rating scales provide for various indicators and behaviors that seem to achieve the necessary reliability and validity. With this article, we hope to stimulate more research on family triads beyond the mother–child dyad. Indeed, the extension of the FA model to adolescence might lead to the possibility of studying families with a more complete perspective, for example, by following birth cohorts from infancy to adolescence within a given framework that allows for a family-level perspective. We hope, ultimately, that future studies using the LTP-CDT and the FCAAS will open new perspectives for developmental research with adolescents. Such work could, in turn, contribute to theoretical advancements in the field.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/adolescents5040052/s1, Document S1: Detailed description of the FCAAS scales; Document S2: Instructions for the LTP-CDT task; Document S3: Mplus output of the simulation study conducted for power analysis; Document S4: Table S1. Spearman correlations between all study and control variables. References [41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67] are cited in the supplementary materials.

Author Contributions

Conceptualization, M.R., N.F. and H.T.; methodology, M.R. and H.T.; software, M.R. and H.T.; formal analysis, M.R. and H.T.; investigation, M.R., A.F., A.B., A.M. and M.S.; data curation, M.R., A.F., A.B., A.M. and M.S.; writing—original draft preparation, M.R., N.F. and H.T.; writing—review and editing, M.R., N.F., A.F., A.B., A.M., M.S. and H.T.; supervision, M.R., N.F. and H.T.; project administration, M.R., N.F. and H.T. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Cantonal Commission on Ethics in Human Research of the State of Vaud (CER-VD; Project n. 2021-01859, 25 October 2021).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data supporting the findings of this study are available at https://doi.org/10.5281/zenodo.15799551.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
CER-VDCommission cantonale d’éthique du canton de Vaud [Cantonal Commission on Ethics in Human Research of the State of Vaud]
CFAConfirmatory factor analysis
CFIComparative fit index
CIPACoparenting inventory for parents and adolescents
FAFamily alliance
FAASFamily alliance assessment scales
FCAASFamily conflict and alliance assessment scales
HROHuman research ordinance
ICCIntraclass correlation coefficient
LTPLausanne trilogue play
LTP-CDTLausanne trilogue play—conflict discussion task
RASRelationship assessment scale
RMSEARoot mean square error of approximation
SEMStructural equation modeling
SRMRStandardized root mean square residual
TLITucker–Lewis Index

References

  1. Van Egeren, L.A.; Hawkins, D.P. Coming to terms with coparenting: Implications of definition and measurement. J. Adult Dev. 2004, 11, 165–178. [Google Scholar] [CrossRef]
  2. Paley, B.; Hajal, N.J. Conceptualizing emotion regulation and coregulation as family-level phenomena. Clin. Child Fam. Psychol. Rev. 2022, 25, 19–43. [Google Scholar] [CrossRef]
  3. Tissot, H.; Lapalus, N.; Frascarolo, F.; Despland, J.-N.; Favez, N. Family alliance in infancy and toddlerhood predicts social cognition in adolescence. J. Child Fam. Stud. 2022, 31, 1338–1349. [Google Scholar] [CrossRef]
  4. Teubert, D.; Pinquart, M. The association between coparenting and child adjustment: A meta-analysis. Parenting 2010, 10, 286–307. [Google Scholar] [CrossRef]
  5. Zhao, F.; Wu, H.; Li, Y.; Zhang, H.; Hou, J. The association between coparenting behavior and internalizing/externalizing problems of children and adolescents: A meta-analysis. Int. J. Environ. Res. Public Health 2022, 19, 10346. [Google Scholar] [CrossRef]
  6. Ronaghan, D.; Gaulke, T.; Theule, J. The association between marital satisfaction and coparenting quality: A meta-analysis. J. Fam. Psychol. 2024, 38, 236–245. [Google Scholar] [CrossRef] [PubMed]
  7. Kerig, P.K.; Lindahl, K.M. Family Observational Coding Systems: Resources for Systemic Research; Psychology Press: Hove, UK, 2000. [Google Scholar]
  8. Grotevant, H.D.; Carlson, C.I. Family interaction coding systems: A descriptive review. Fam. Process 1987, 26, 49–74. [Google Scholar] [CrossRef]
  9. Hampson, R.B.; Beavers, W.R.; Hulgus, Y.F. Insiders’ and outsiders’ views of family: The assessment of family competence and style. J. Fam. Psychol. 1989, 3, 118–136. [Google Scholar] [CrossRef]
  10. Mollà Cusí, L.; Günther-Bel, C.; Vilaregut Puigdesens, A.; Campreciós Orriols, M.; Matalí Costa, J.L. Instruments for the assessment of coparenting: A systematic review. J. Child Fam. Stud. 2020, 29, 2487–2506. [Google Scholar] [CrossRef]
  11. Baril, M.E.; Crouter, A.C.; McHale, S.M. Processes linking adolescent well-being, marital love, and coparenting. J. Fam. Psychol. 2007, 21, 645–654. [Google Scholar] [CrossRef]
  12. Fivaz-Depeursinge, E.; Corboz-Warnery, A. The Primary Triangle: A Developmental Systems View of Mothers, Fathers, and Infants; Basic Books: New York, NY, USA, 1999. [Google Scholar]
  13. Tissot, H.; Favez, N. The Lausanne Trilogue Play: Bringing together developmental and systemic perspectives in clinical settings. Aust. N. Z. J. Fam. Ther. 2023, 44, 511–520. [Google Scholar] [CrossRef]
  14. Favez, N.; Lavanchy Scaiola, C.; Tissot, H.; Darwiche, J.; Frascarolo, F. The family alliance assessment scales: Steps toward validity and reliability of an observational assessment tool for early family interactions. J. Child Fam. Stud. 2011, 20, 23–37. [Google Scholar] [CrossRef]
  15. Favez, N.; Lopes, F.; Bernard, M.; Frascarolo, F.; Lavanchy Scaiola, C.; Corboz-Warnery, A.; Fivaz-Depeursinge, E. The development of family alliance from pregnancy to toddlerhood and child outcomes at 5 years. Fam. Process 2012, 51, 542–556. [Google Scholar] [CrossRef]
  16. Favez, N.; Frascarolo, F.; Carneiro, C.; Montfort, V.; Corboz-Warnery, A.; Fivaz-Depeursinge, E. The development of the family alliance from pregnancy to toddlerhood and children outcomes at 18 months. Infant Child Dev. 2006, 15, 59–73. [Google Scholar] [CrossRef]
  17. Mensi, M.M.; Orlandi, M.; Rogantini, C.; Provenzi, L.; Chiappedi, M.; Criscuolo, M.; Castiglioni, M.C.; Zanna, V.; Borgatti, R. Assessing family functioning before and after an integrated multidisciplinary family treatment for adolescents with restrictive eating disorders. Front. Psychiatry 2021, 12, 693. [Google Scholar] [CrossRef] [PubMed]
  18. Gottman, J.M.; Notarius, C.I. Decade review: Observing marital interaction. J. Marriage Fam. 2000, 62, 927–947. [Google Scholar] [CrossRef]
  19. Persram, R.J.; Scirocco, A.; Della Porta, S.; Howe, N. Moving beyond the dyad: Broadening our understanding of family conflict. Hum. Dev. 2019, 63, 38–70. [Google Scholar] [CrossRef]
  20. Branje, S. Development of parent–adolescent relationships: Conflict interactions as a mechanism of change. Child Dev. Perspect. 2018, 12, 171–176. [Google Scholar] [CrossRef]
  21. Warmuth, K.A.; Cummings, E.M.; Davies, P.T. Constructive and destructive interparental conflict, problematic parenting practices, and children’s symptoms of psychopathology. J. Fam. Psychol. 2020, 34, 301–311. [Google Scholar] [CrossRef]
  22. Weymouth, B.B.; Buehler, C.; Zhou, N.; Henson, R.A. A meta-analysis of parent–adolescent conflict: Disagreement, hostility, and youth maladjustment. J. Fam. Theory Rev. 2016, 8, 95–112. [Google Scholar] [CrossRef]
  23. Romet, M.; Favez, N.; Tissot, H. Family Conflict and Alliance Assessment Scales—With Adolescents (FCAAS); FCAAS: Foxborough, MA, USA; Lausanne, Switzerland, 2023; unpublished. [Google Scholar]
  24. Harris, P.A.; Taylor, R.; Thielke, R.; Payne, J.; Gonzalez, N.; Conde, J.G. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. J. Biomed. Inform. 2009, 42, 377–381. [Google Scholar] [CrossRef]
  25. Hendrick, S.S. A generic measure of relationship satisfaction. J. Marriage Fam. 1988, 50, 93–98. [Google Scholar] [CrossRef]
  26. Saramago, M.; Lemétayer, F.; Gana, K. Adaptation et validation de la version française de l’échelle d’évaluation de la relation. Psychol. Française 2021, 66, 333–343. [Google Scholar] [CrossRef]
  27. Teubert, D.; Pinquart, M. The Coparenting Inventory for Parents and Adolescents (CI-PA): Reliability and validity. Eur. J. Psychol. Assess. 2011, 27, 206–215. [Google Scholar] [CrossRef]
  28. Zimmermann, G.; Antonietti, J.-P.; Sznitman, G.A.; Petegem, S.V.; Darwiche, J. The French version of the coparenting inventory for parents and adolescents (CI-PA): Psychometric properties and a cluster analytic approach. J. Fam. Stud. 2020, 28, 652–677. [Google Scholar] [CrossRef]
  29. Favez, N.; Tissot, H.; Frascarolo, F. Is it typical? The ecological validity of the observation of mother-father-infant interactions in the Lausanne Trilogue Play. Eur. J. Dev. Psychol. 2019, 16, 113–121. [Google Scholar] [CrossRef]
  30. Ten Hove, D.; Jorgensen, T.D.; van der Ark, L.A. Updated guidelines on selecting an intraclass correlation coefficient for interrater reliability, with applications to incomplete observational designs. Psychol. Methods 2022. [Google Scholar] [CrossRef] [PubMed]
  31. Muthén, L.K.; Muthén, B.O. Mplus User’s Guide, 8th ed.; Muthén & Muthén: Los Angeles, CA, USA, 1998–2023. [Google Scholar]
  32. Hu, L.-t.; Bentler, P.M. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct. Equ. Model. A Multidiscip. J. 1999, 6, 1–55. [Google Scholar] [CrossRef]
  33. Whittaker, T.A. Using the modification index and standardized expected parameter change for model modification. J. Exp. Educ. 2012, 80, 26–44. [Google Scholar] [CrossRef]
  34. Chen, F.F. Sensitivity of goodness of fit indexes to lack of measurement invariance. Struct. Equ. Model. A Multidiscip. J. 2007, 14, 464–504. [Google Scholar] [CrossRef]
  35. Cheung, G.W.; Rensvold, R.B. Evaluating goodness-of-fit indexes for testing measurement invariance. Struct. Equ. Model. A Multidiscip. J. 2002, 9, 233–255. [Google Scholar] [CrossRef]
  36. Minuchin, S. Families & Family Therapy; Harvard University Press: Boston, MA, USA, 1974. [Google Scholar]
  37. Ece, C.; Gürmen, M.S.; Acar, İ.H.; Buyukcan-Tetik, A. Examining the dyadic association between marital satisfaction and coparenting of parents with young children. Curr. Psychol. 2023, 43, 1473–1482. [Google Scholar] [CrossRef]
  38. Beam, C.R.; Marcus, K.; Turkheimer, E.; Emery, R.E. Gender differences in the structure of marital quality. Behav. Genet. 2018, 48, 209–223. [Google Scholar] [CrossRef]
  39. Hetherington, E.M. Family functioning and the adjustment of adolescent siblings in diverse types of families. Monogr. Soc. Res. Child Dev. 1999, 64, 1–25. [Google Scholar] [CrossRef] [PubMed]
  40. McHale, J.; Tissot, H.; Mazzoni, S.; Hedenbro, M.; Salman-Engin, S.; Philipp, D.A.; Darwiche, J.; Keren, M.; Collins, R.; Coates, E.; et al. Framing the work: A coparenting model for guiding infant mental health engagement with families. Infant Ment. Health J. 2023, 44, 638–650. [Google Scholar] [CrossRef] [PubMed]
  41. Goodwin, C. Restarts, pauses, and the achievement of a state of mutual gaze at turn-beginning. Sociol. Inq. 1980, 50, 272–302. [Google Scholar] [CrossRef]
  42. Goodwin, C. Conversational Organization: Interaction Between Speakers and Hearers; Academic Press: New York, NY, USA, 1981. [Google Scholar]
  43. Robinson, J.D. Getting down to business: Talk, gaze, and body orientation during openings of doctor-patient consultations. Hum. Commun. Res. 1998, 25, 97–123. [Google Scholar] [CrossRef]
  44. Beveridge, R.M.; Berg, C.A. Parent–adolescent collaboration: An interpersonal model for understanding optimal interactions. Clin. Child Fam. Psychol. Rev. 2007, 10, 25–52. [Google Scholar] [CrossRef]
  45. Favez, N.; Frascarolo, F.; Tissot, H. The family alliance model: A way to study and characterize early family interactions. Front. Psychol. 2017, 8, 1441. [Google Scholar] [CrossRef]
  46. Coan, J.A.; Gottman, J.M. The specific affect coding system (SPAFF). In Handbook of Emotion Elicitation and Assessment; Coan, J.A., Allen, J.J.B., Eds.; Oxford University Press: Oxford, UK, 2007; pp. 267–285. [Google Scholar]
  47. Bell, D.C.; Bell, L.G. Micro and macro measurement of family systems concepts. J. Fam. Psychol. 1989, 3, 137–157. [Google Scholar] [CrossRef]
  48. Lee, K. Why Do We Overlap Each Other?: Collaborative Overlapping Talk in English as a Lingua Franca (ELF) Communication. Korean J. Engl. Lang. Linguist. 2020, 20, 613–641. [Google Scholar]
  49. Available online: https://enfance-et-partage.org/wp-content/uploads/2019/05/guide-parentalite-version-def.pdf (accessed on 18 September 2025).
  50. Forbes, C.; Vuchinich, S.; Kneedler, B. Assessing families with the family problem solving code. In Family Observational Coding Systems: Resources for Systemic Research; Kerig, P., Lindahl, K., Eds.; Psychology Press: Hove, UK, 2000; pp. 59–75. [Google Scholar]
  51. Marceau, K.; Zahn-Waxler, C.; Shirtcliff, E.A.; Schreiber, J.E.; Hastings, P.; Klimes-Dougan, B. Adolescents’, mothers’, and fathers’ gendered coping strategies during conflict: Youth and parent influences on conflict resolution and psychopathology. Dev. Psychopathol. 2015, 27, 1025–1044. [Google Scholar] [CrossRef]
  52. Baron, R.A. Negative effects of destructive criticism: Impact on conflict, self-efficacy, and task performance. J. Appl. Psychol. 1988, 73, 199–207. [Google Scholar] [CrossRef]
  53. Fonagy, P.; Steele, M.; Steele, H.; Moran, G.S.; Higgitt, A.C. The capacity for understanding mental states: The reflective self in parent and child and its significance for security of attachment. Infant Ment. Health J. 1991, 12, 201–218. [Google Scholar] [CrossRef]
  54. Vanwoerden, S. The Development and Validation of an Observational Coding System for Real-Time Parent-Adolescent Mentalizing. Ph.D. Dissertation, University of Houston, Houston, TX, USA, 2020. [Google Scholar]
  55. Ballespí, S.; Vives, J.; Debbané, M.; Sharp, C.; Barrantes-Vidal, N. Beyond diagnosis: Mentalization and mental health from a transdiagnostic point of view in adolescents from non-clinical population. Psychiatry Res. 2018, 270, 755–763. [Google Scholar] [CrossRef]
  56. Vuchinich, S.; Vuchinich, R.; Wood, B. The Interparental Relationship and Family Problem Solving with Preadolescent Males. Child Dev. 1993, 64, 1389–1400. [Google Scholar] [CrossRef] [PubMed]
  57. Doane, J.A. Family Interaction and Communication Deviance in Disturbed and Normal Families: A Review of Research. Fam. Process. 1978, 17, 357–376. [Google Scholar] [CrossRef] [PubMed]
  58. Brotherton, W.D. The Assessment of Parental Triangulation of Children; The Florida State University: Tallahassee, FL, USA, 1989. [Google Scholar]
  59. Kerr, M.E.; Bowen, M. Family Evaluation; WW Norton & Company: New York, NY, USA, 1988. [Google Scholar]
  60. Minuchin, S.; Baker, L.; Rosman, B.L.; Liebman, R.; Milman, L.; Todd, T.C. A conceptual model of psychosomatic illness in children: Family organization and family therapy. Arch. Gen. Psychiatry 1975, 32, 1031–1038. [Google Scholar] [CrossRef]
  61. Allen, J.P.; Hauser, S.T.; Bell, K.L.; O’Connor, T.G. Longitudinal assessment of autonomy and relatedness in adolescent-family interactions as predictors of adolescent ego development and self-esteem. Child Dev. 1994, 65, 179–194. [Google Scholar] [CrossRef] [PubMed]
  62. Allen, J.P.; Hauser, S.T. Autonomy and relatedness in adolescent-family interactions as predictors of young adults’ states of mind regarding attachment. Dev. Psychopathol. 1996, 8, 793–809. [Google Scholar] [CrossRef]
  63. Bengtson, P.L.; Grotevant, H.D. The individuality and connectedness Q-sort: A measure for assessing individuality and con-nectedness in dyadic relationships. Pers. Relatsh. 1999, 6, 213–225. [Google Scholar] [CrossRef]
  64. Chaplin, T.M.; Sinha, R.; Simmons, J.A.; Healy, S.M.; Mayes, L.C.; Hommer, R.E.; Crowley, M.J. Parent–adolescent conflict interactions and adolescent alcohol use. Addict. Behav. 2012, 37, 605–612. [Google Scholar] [CrossRef]
  65. Maharaj, S.; Rodin, G.; Connolly, J.; Olmsted, M.; Daneman, D. Eating problems and the observed quality of mother–daughter interactions among girls with type 1 diabetes. J. Consult. Clin. Psychol. 2001, 69, 950–958. [Google Scholar] [CrossRef] [PubMed]
  66. Ravindran, N.; Hu, Y.; McElwain, N.L.; Telzer, E.H. Dynamics of mother–adolescent and father–adolescent autonomy and control during a conflict discussion task. J. Fam. Psychol. 2020, 34, 312–321. [Google Scholar] [CrossRef] [PubMed]
  67. Van Petegem, S.; Baudat, S.; Zimmermann, G. Interdit d’interdire? Vers une meilleure compréhension de l’autonomie et des règles au sein des relations parents-adolescents. Can. Psychol. 2019, 60, 194–202. [Google Scholar] [CrossRef]
Adolescents 05 00052 g001
Figure 1. Standardized parameter estimates of the confirmatory two-factor model. All indicators and residual variances were highly significant (p < 0.01), except for the residual variance of the scale “Role reversal” (p = 0.118). FA = Family Alliance; COPA = Coparenting.
Figure 1. Standardized parameter estimates of the confirmatory two-factor model. All indicators and residual variances were highly significant (p < 0.01), except for the residual variance of the scale “Role reversal” (p = 0.118). FA = Family Alliance; COPA = Coparenting.
Adolescents 05 00052 g001
Table 1. FCAAS scales—summary.
Table 1. FCAAS scales—summary.
Rating ScalesBrief Description of Appropriate Criteria
Postures & gazesAt the non-verbal level, participants signal their availability to interact, and everyone is involved in the interaction
Turn-takingAt the verbal level, everyone is engaged, talk time is balanced and there are few interruptions or monologues
Mutual respect for coparenting rolesBoth parents are compliant with the LTP-CDT instructions regarding the various interactive roles (third-party or active)
Conflict resolutionThanks to cooperation in the family, a negotiation process allows for the problem to be solved through dialogue and co-construction of a viable solution
Affective climateThe emotional climate is positive and warm, while affects are authentic; family members seem to enjoy each other’s presence
MentalizationFamily members pay attention to their own and others’ mental states; there is a climate of validation and empathy
Role reversalThe adolescent is not involved in the parental subsystem
Coparenting supportParents agree about education or at least they coordinate and support each other
Autonomy promotionParents show respect for the adolescent’s individuality, and help them identify and express their needs and preferences; limit management is clear and flexible
Adolescent autonomyThe adolescent demonstrates autonomy, independence, and self-approval
Note. The score ranges from 1 (inappropriate) to 5 (appropriate), with scores of 3 representing the moderate range. Scores of 2 and 4 allow us to nuance the ratings.
Table 2. ICCs and variance components of the FCAAS scales.
Table 2. ICCs and variance components of the FCAAS scales.
FCAAS ScalesICCLower BoundUpper BoundSE
Postures & Gazes0.640.330.900.13
Turn-taking0.830.680.920.05
Mutual respect for coparenting roles0.820.660.970.08
Conflict resolution0.900.810.960.03
Affective climate0.860.750.940.05
Mentalization0.880.780.950.04
Role reversal0.900.820.960.03
Coparenting support0.850.730.940.05
Autonomy promotion0.860.730.940.05
Adolescent autonomy0.810.640.920.06
Note. N = 82.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Romet, M.; Favez, N.; Foletta, A.; Burnier, A.; Mrozek, A.; Schumacher, M.; Tissot, H. Presentation and Initial Validation of a New Observational Situation and Coding System for Assessing Triadic Family Interactions with Adolescents. Adolescents 2025, 5, 52. https://doi.org/10.3390/adolescents5040052

AMA Style

Romet M, Favez N, Foletta A, Burnier A, Mrozek A, Schumacher M, Tissot H. Presentation and Initial Validation of a New Observational Situation and Coding System for Assessing Triadic Family Interactions with Adolescents. Adolescents. 2025; 5(4):52. https://doi.org/10.3390/adolescents5040052

Chicago/Turabian Style

Romet, Michaël, Nicolas Favez, Amalia Foletta, Annie Burnier, Aleksandra Mrozek, Marie Schumacher, and Hervé Tissot. 2025. "Presentation and Initial Validation of a New Observational Situation and Coding System for Assessing Triadic Family Interactions with Adolescents" Adolescents 5, no. 4: 52. https://doi.org/10.3390/adolescents5040052

APA Style

Romet, M., Favez, N., Foletta, A., Burnier, A., Mrozek, A., Schumacher, M., & Tissot, H. (2025). Presentation and Initial Validation of a New Observational Situation and Coding System for Assessing Triadic Family Interactions with Adolescents. Adolescents, 5(4), 52. https://doi.org/10.3390/adolescents5040052

Article Metrics

Back to TopTop