EMOTIONAL LIFE OF THE FAMILY: PARENTAL META-EMOTIONS, CHILDREN'S TEMPERAMENT AND INTERNALISING AND EXTERNALISING PROBLEMS

Abstract

The aim of this study was to investigate some assumptions related to Gottman's model of parental meta-emotions on a sample of preschool children's parents. The study included 506 parents. The questionnaires about parental meta-emotions, dimensions of temperament and children's internalising and externalising problems were completed by mothers and fathers separately. Mothers' and fathers' awareness of their own and of their children's emotions and coaching child emotions were indirectly related to child externalising and internalising problems through child negative affectivity and effortful control, and awareness was also directly related to the child's externalising and internalising problems. The results of the models showed that mothers' and fathers' awareness and coaching were positively related to effortful control and negatively related to negative affectivity, whereas only mothers' and fathers' awareness were negatively related to externalising and internalising problems. Simultaneously, negative affectivity was positively related to the child's externalising and internalising problems and effortful control was negatively related to the child's externalising and internalising problems. With this study, we have confirmed part of the investigated assumptions from Gottman's model that relate to the influence of parental meta-emotion on child outcomes.

Headnote

The aim of this study was to investigate some assumptions related to Gottman's model of parental meta-emotions on a sample of preschool children's parents. The study included 506 parents. The questionnaires about parental meta-emotions, dimensions of temperament and children's internalising and externalising problems were completed by mothers and fathers separately. Mothers' and fathers' awareness of their own and of their children's emotions and coaching child emotions were indirectly related to child externalising and internalising problems through child negative affectivity and effortful control, and awareness was also directly related to the child's externalising and internalising problems. The results of the models showed that mothers' and fathers' awareness and coaching were positively related to effortful control and negatively related to negative affectivity, whereas only mothers' and fathers' awareness were negatively related to externalising and internalising problems. Simultaneously, negative affectivity was positively related to the child's externalising and internalising problems and effortful control was negatively related to the child's externalising and internalising problems. With this study, we have confirmed part of the investigated assumptions from Gottman's model that relate to the influence of parental meta-emotion on child outcomes.

Keywords: parental meta-emotion, negative affectivity, effortful control, child externalising and internalising problems

Emocionalni zivot obitelji: roditeljske metaemocije, temperament djeteta i eksternalizirani i internalizirani problemi

Andreja BRAJSA-ZGANEC

Institut drustvenih znanosti Ivo Pilar, Zagreb

Cilj istrazivanja bio je ispitati neke pretpostavke Gottmanova modela roditeljskih metaemocija na uzorku roditelja djece predskolske dobi. U istrazivanju je sudjelovalo 506 roditelja. Upitnike o metaemocijama, dimenzijama temperamenta te djecjim internaliziranim i eksternaliziranim problemima ispunjavali su zasebno majka i otac. Majcina i oceva svjesnost vlastitih i djecjih emocija i postupanje s djecjim emocijama indirektno je povezana s djecjim eksternali- ziranim i internaliziranim problemima preko djecje negativne afektivnosti i samokontrole a dodatno je svjesnost izravno povezana s djecjim internaliziranim i eksternaliziranim problemima. Rezultati dobivenih modela pokazuju da je majcina i oceva svjesnost vlastitih i djecjih emocija i postupanje s djecjim emocijama pozitivno povezana sa samokontrolom i negativno povezana s negativnom afektivnosti, dok je jedino majcina i oceva svjesnost vlastitih i djecjih emocija negativno povezana s eksternaliziranim i internaliziranim problemima. Istodobno, negativna je afektivnost pozitivno povezana s djecjim eksternaliziranim i internaliziranim problemima, a samokontrola je negativno povezana s djecjim eksternaliziranim i internaliziranim problemima. Ovim istrazivanjem potvrdili smo dio pretpostavki iz Gottmanova modela kako roditeljske metaemocije utjecu na djecje izlazne varijable.

Kljucne rijeci: roditeljske metaemocije, negativna afektivnost, samokontrola, djecji eksternalizirani i internalizirani problemi

INTRODUCTION

The family is one of the main socialization contexts of emo- tional expression and regulation among normative samples (Denham, 1997; Denham et al., 2000; Eisenberg, Fabes, & Lo- soya, 1997; Eisenberg et al., 1995; Eisenberg et al., 2001). Among recent approaches used to investigate the relationship between family and children's adjustment are those oriented toward the structure of parental meta-emotions and the functioning role of meta-emotions in children's development (Gottman, Katz, & Hooven, 1996; Hooven, Gottman, & Fainsilber Katz, 1995; Katz, Maliken, & Stettler, 2012). Gottman and collabora- tors emphasize the importance of parental meta-emotion struc- ture for maintaining the family as a system. In their research, Gottman and collaborators investigate parental awareness of their own and of their children's emotions and parents' abili- ty to manage children's emotions as important predictors of children's psychosocial adjustment (Gottman et al., 1996; Hoo- ven et al., 1995; Katz & Gottman, 1997). The focus of their later research is on the association between specific selected di- mensions of meta-emotions and problems in childhood (Katz & Windecker-Nelson, 2004; Katz & Windecker-Nelson, 2006; Maliken & Katz, 2013; Ramsden & Hubbard, 2002) and in ado- lescence (Katz & Hunter, 2007; Wu Shortt, Stoolmiller, Smith- -Shine, Eddy, & Sheeber, 2010; Hunter et al., 2011).

The concept of meta-emotions refers to emotions about emotions, to the analysis of our own emotions. For Gottman and collaborators, the meta-emotion structure implies emotional executive functioning based on emotions; it includes thoughts and emotions about emotions where meta-emotion encom- passes thoughts, metaphors, and an entire parental philosophy about emotions. Using the meta-emotion concept, Gottman and collaborators (1996, 1997), in their research, evaluated pa- rents' emotions about their own emotions in order to reveal in- dividual differences in the experience of emotions. People react differently when they experience anger; some are visi- bly upset when they get angry, some feel well in relation to their ability to express anger and there are also those who are neutral about their anger - neither good nor bad. In addition, the approach to sadness differs among individuals. The con- cepts parallel to meta-emotions are meta-cognition and meta- communication (Gottman et al., 1997; Hooven et al., 1995; Katz & Gottman, 1997). Another concept comparable to that of me- ta-emotions is meta-mood; meta-experience about mood which involves cognition and an opinion of an individual about his own mood. These concepts, similar to meta-emotions, have been discussed and investigated by theorists and researchers in the field of emotional intelligence. Mayer and Salovey (1997), in their model of emotional intelligence, identify three levels of emotional intelligence; perception and expression of our emo- tions and among others, the use of emotions for better adap- tation and our emotion regulation and emotion regulation in others. According to the theory of emotional intelligence, people differ in awareness and regulation of their own emotions, in their ability to manage their own and others' emotions and in their ability to reflexively track emotions related to ourselves and to others.

Comparable to the concept of emotional intelligence, a- wareness of one's own and children's emotions and coaching of emotions are incorporated in the concept of meta-emotion structure. According to Gottman and collaborators, the par- ent-child interaction related to child's emotions takes place through parents' actions and communication with the child about his/her emotions, about the conditions that cause them and strategies to deal with emotions. In their research, Got- tman and collaborators confirmed the existence of the con- cept of meta-emotions and its multiple associations with psy- chosocial adjustment in children (Gottman et al., 1996, 1997; Hooven et al., 1995; Katz & Gottman, 1997).

There is much research in developmental psychology that relates to parents' actions and discipline, variables such as warmth, control, authoritarian or authoritative style (see Baumrid, 1971; Maccoby & Martin, 1983; Darling & Steinberg, 1993; Lansford, Wager, Bates, Dodge, & Pettit, 2012 for a re- view). Considerably less research has focused on parents' coa- ching their children's emotions and to emotional bonds be- tween parents and children. According to Gottman and col- laborators, what is missing in the described parental styles are parents' emotions about their own emotions and connections with specific emotional behaviours in children. Meta-emotions and teaching about emotions is not a sub-dimension of posi- tive parenting but an additional component to the current concept of parenthood and is somewhat more general than the concept. According to Gottman and collaborators (1996, 1997), parents' relationship toward children's emotional life is associated with children's regulation of emotions and influ- ences children's behaviour in everyday life.

The starting point of the model by Gottman, Katz, and Hooven depicting the philosophy of parents' meta-emotions and emotional family life, is the association between parents' emotions and thoughts on their emotions and decreased pa- rental negative affect and increased positive parenting as well as the assumption that parental meta-emotion structure direct- ly influences children's regulatory physiology and children's ability to regulate their emotions. In this way, the parental meta-emotion structure is also connected with children's psy- chosocial adjustment. A basic hypothesis of the model is par- ents' awareness of the emotional life of their child and par- ents' attempts to establish emotional ties with the child. In their research, Gottman and collaborators (1996, 1997) started from a total of 12 variables; awareness of own emotions, aware- ness of children's emotions and coaching children's emotions for mothers and fathers; sadness and anger variables that were reduced to two variables of parental meta-emotion structure, such as awareness of own emotions and awareness of chil- dren's emotions and coaching children's emotions. Some ana- lyses dealt with the meta-emotion structure separately for mothers and fathers (Hooven et al., 1995) and dimensions of awareness and coaching were summarised (Katz & Gottman, 1997) in some analyses.

Individual differences in children's emotionality are close- ly related to their regulatory physiology and emotion regula- tion and therefore play an important role in the influence that meta-emotions have on emotional and social competence as well as emotional or behavioural problems in children, given that, as assumed by the authors, parents coach children's emotions in compliance to the characteristics of the child's temperament. In a great deal of research, parental discipline methods are associated with temperament dimensions and with psychosocial adjustment measures (e.g. Stright, Galla- gher, & Kelley, 2008; Russell, Hart, Robinson, & Olsen, 2003; van Zeijl et al., 2007; Brajsa-Zganec & Hanzec, 2014). In the mo- del of meta-emotions, characteristics of temperament are re- lated to the parenting style that reduces the importance of emotions and negative relations with peers (Gottman et al., 1996). Parents' meta-emotions may be influenced by the child's temperament characteristics but parents' coaching of chil- dren's emotions in interaction with children's temperament characteristics, may influence children's psychosocial adjust- ment (Gottman et al., 1997; Katz, Gottman, & Hooven, 1996).

Generally, results of testing Gottman's model pointed to a direct connection between meta-emotion variables and chil- dren's psychosocial adjustment, children's health in particular, and also to indirect connections through children's physiolo- gy and emotion regulation for the criteria variable of conduct towards peers, and through the styles of parental behaviour especially in relation to children's achievement.

Since the model by Gottman and collaborators deals with emotional relations within the family, as well as with chil- dren's emotional and social competency and children's psy- chosocial adjustment, the model has been criticised by resear- chers in both fields. Nancy Eisenberg (1996) emphasised the overlapping of the parent's meta-emotion with parental style variables and suggested that the model should pay greater variable attention to children's emotion regulation and chil- dren's emotionality. Similarly, Philip Cowan (1996) empha- sised the problem of distinction between meta-emotion struc- ture with parental styles and the importance of direct influ- ence of parents' awareness of their own and child's emotions and coaching the child's emotions on children's psychosocial adjustment.

Our interest was to investigate some assumptions of this model among a Croatian sample of parents with preschool children. The aim of our study was to assess the relations be- tween family emotional variables, mothers' and fathers' a- wareness of their own and their child's emotions and coach- ing children's emotion and negative affectivity and effortful control as dimensions of temperament as well as between mo- thers' awareness and coaching of emotions and externalising and internalising problems of children. Our intention was to check, on the part of Gottman's model relating to parental me- ta-emotions, children's temperament and internalising and externalising problems. On the basis of this model, we hy- pothesised that there is a direct association between parental meta-emotions and children's problems as well as their indi- rect associations through the dimension of the child's tem- perament.

METHOD

Participants

The study included 506 parents from complete families. The completed surveys given to parents were returned by 556 mothers and fathers for a total of 716 children, 97% (538) of parents were married and living together. All the required in- formation was collected from mothers and fathers for 506 (94%) children. The mean age of preschool children in our sample was 5 years and 2 months; the age range was from 2 years and 9 months to 7 years and 3 months. In our final sample, 24% were single children, 58% had a brother or sister, and 18% were from families with three or more children. Parents' edu- cation (1 - elementary school not completed; 2 - elementary school; 3 - secondary school; 4 - non-university post-sec- ondary education; 5 - university or higher level) was the fol- lowing: on average, mothers (57%) and fathers (61%) com- pleted secondary school and approximately an equal number of mothers and fathers had a non-university post-secondary (13%) or university level of education (22%).

Measures

Parental meta-emotion structure

Mothers and fathers completed the new Parental Meta-Emo- tion Structure Questionnaire (Brajsa-Zganec, 2002) construct- ed and based on the theoretical hypotheses of the model for parental meta-emotion structure (Gottman et al., 1996, 1997), Meta-Emotion Interview (Katz & Gottman, 1986) and coding system for the Meta-Emotion Interview (Gottman et al., 1996; Hooven et al., 1995). Two preliminary studies were conduct- ed among students (see Brajsa-Zganec, 2003). A final selection of statements was made using a series of analyses conducted on a sample of parents from the main study. In subsequent re- -analyses the scales were additionally shortened and were cross- -validated again on a new sample of subjects from the same population and all analyses were conducted separately for fathers and mothers (see Brajsa-Zganec, 2003).

In this study, the overall result for two measures of meta- -emotional structures related to both mothers and fathers. A dimension of awareness of their own and children's emotions of sadness and intense anger was assessed using 26 items, and a dimension of coaching children's emotions using 10 items in the questionnaire. Theoretical results on each of the two measures of dimensions for meta-emotions ranged from 1 to 4 because the average result calculated for each measure was expressed in scale units. Statements relating to lower a- wareness of parents' own and children's emotions of sadness and intense anger as well as emotion coaching were recoded in such a way that a higher score indicated higher awareness of their own and children's emotions of sadness and intense anger, and also coaching based on greater respect, that is, bet- ter coaching children's emotions of sadness and intense anger.

For dimensions of awareness of their own and children's emotions of sadness and intense anger, the reliability coeffi- cient (Cronbach alpha) is 0.85 for mothers, and 0.83 for fa- thers. For dimensions of coaching children's emotions of sad- ness and intense anger, the reliability coefficient is 0.69 for mothers and 0.74 for fathers.

Children's externalising and internalising problems

For the assessment of children's problems, mothers and fa- thers completed Achenbach's Child Behavior Checklist/ 4-18 (CBCL, Achenbach, 1991). The latent structure of behaviour problems and other metric characteristics were tested and the obtained results were used to create subscales of problem behaviour (see Brajsa-Zganec, 2003). CBCL was filled in by the mother and father for each child and on the basis of a fac- torial structure two subscales were created. The subscale for the externalised problems consists of 22 items while for the internalised problems this consists of 13 items.

The average assessment value was calculated for moth- ers and fathers. Justification of composite measures for the assessment of children's problems by both parents was car- ried out by studying the convergent and discriminative valid- ity of these measures using a multitrait-multimethod analysis by Campbell and Fiske (1959). Although all coefficients of convergent and discriminative validity were found signifi- cant, they indicate a satisfactory convergent and discrimina- tive validity of measures and justify the formation of a com- posite result for externalized and internalized problems for each child for both parents. Coefficients of convergent valid- ity are higher than those of discriminative validity. The total value for externalised problems and internalised problems is in the range from 0 to 44 and from 0 to 26, respectively. Relia- bility (Cronbach alpha coefficients) of composite results ob- tained on the basis of average assessments by both parents is 0.90 for externalised and 0.80 for internalised problems. Re- liability values are somewhat higher than those for individual assessments and indicate satisfactory homogenous composite measures for children's behaviour.

Children's temperament questionnaire

In this study, a part of the Children's Behaviour Question- naire was used for parental assessment of children's tempera- ment characteristics (CBQ, Rothbart, 1988; according to Gold- smith & Rothbart, 1991). We examined two dimensions of temperament; effortful control and negative affectivity, and each of them was measured using two subscales. The basic di- mensions included in the CBQ for the child's first year (Gold- smith & Rothbart, 1991) had inhibitory control added as an important aspect of socialisation, social development and pro- blematic behaviour (Kochanska, Murray, Jacques, Koenig, & Vandegeest, 1996; Kochanska, Murray, & Coy, 1997; Kochan- ska, Murray, & Harlan, 2000). Abbreviated CBQ in this inves- tigation included the following scales: fear, frustration, atten- tion focus and inhibitory control, and two characteristics for temperament of two higher-order dimensions each (negative affectivity and inhibitory control). Parents were asked to esti- mate to what degree each statement describes their child's response over the past six months and to mark their estimate on a scale from 1 to 7 (1 - extremely untrue of your child, 2 - quite untrue, 3 - slightly untrue, 4 - neither true nor false, 5 - slight- ly true, 6 - quite true, 7 - extremely true of your child or NA meaning "Non applicable to the child"). Frustration, inhibito- ry control and attention focus consists of 11 items and the fear scale consists of 10 items. As suggested in the studies of tem- perament (Ahadi, Rothbart, & Ye, 1993; Rothbart, Ahadi, & Hershey, 1994) and on basis of the analyses conducted, as- sessments for basic dimensions of temperament for mothers and fathers were created and the average result was calculat- ed for each higher-order dimension. Possible theoretical re- sults obtained in this way for each of the higher-order di- mensions are in the range from 1 to 7 for both mothers and fathers, where a higher score denotes manifestation of behav- iour more indicative of a certain characteristic of temperament. As suggested by researchers in this field (e.g. Dettling, Gun- nar, & Donzella, 1999; Eisenberg & Fabes, 1995; Katz & Got- tman, 1993) composite measures were created for mothers and fathers for the two measures of temperament dimensions and their convergent and discriminative validity were studied (see Brajsa-Zganec, 2003). The conducted analyses justify the creation of composite results for the two dimensions of tem- perament for each child as the average assessment value by the mother and father. Reliability (Cronbach alpha coeffici- ents) of composite results obtained on the basis of average assessments by both parents range from 0.77 to 0.87, they are somewhat higher than the reliability of individual estimates and indicate satisfactory homogeneity of composite measures of temperament dimensions.

Procedure

The dataset used in this paper is a part of a larger dataset col- lected in a research project conducted in Croatia that focused on meta-emotion and family, on socio-emotional develop- ment of preschool children and on children's basic behaviou- ral characteristics (Brajsa-Zganec, 2003). Approval for this pro- ject conducted in kindergartens was provided from the Mini- stry of Education, Science and Sport of the Republic of Croatia and kindergarten principals. The study included 37 kinder- garten groups. Parents were given two envelopes containing surveys about their children by their child's teacher; one for the mother and the other for the father. They were asked to fill in the survey at home and after the entire questionnaire was completed by both parents, they were asked to seal the envelopes and return them to teachers within seven days.

RESULTS

Descriptive and correlation analyses

Table 1 presents the means, standard deviations and correla- tions between all the study variables. Gender differences in preschool children were checked for all variables studied, but none of the differences were statistically significant. The age of children was significantly related only to effortful control and child externalising problems. Table 1 shows correlations between dimensions of meta-emotion for the mother and father, negative affectivity and effortful control as a dimen- sion of the child's temperament and the child's externalising and internalising problems. Mothers' and fathers' awareness of their own and their child's emotions were significantly cor- related. In addition, but to a somewhat lesser degree, moth- ers' and fathers' coaching of the child's emotions were signif- icantly correlated. A certain degree of concordance between parents was observed in families relating to parents' aware- ness of their own and their child's emotions, and in the coach- ing of a child's emotion. Conversely, a zero-order correlation for mothers and fathers between awareness of their own and their child's emotions and the coaching of a child's emotion, yielded no significant relationships. Negative affectivity and effortful control were significantly correlated in such a way that children with more effortful control have less negative af- fectivity and vice versa. Mothers' awareness of their own and their child's emotions and their coaching of a child's emotions, and fathers' awareness of their own and their child's emotions were significantly negatively correlated with negative affecti- vity, and mothers' and fathers' awareness and coaching were significantly and positively correlated with effortful control.

Mothers' awareness of their own and their child's emo- tions and fathers' awareness of their own and their child's emo- tions were significantly negatively correlated with children's externalising and internalising problems. These correlations indicate that more mothers' and fathers' awareness of their own and their child's emotion were related to less children's externalising and internalising problems, while mothers' and fathers' coaching of the child's emotions was not significantly correlated with children's externalising and internalising pro- blems. In addition, more effortful control was related to less children's externalising and internalising problems and more negative affectivity was related to more children's externalis- ing and internalising problems.

Path analyses

Using correlations described above and the model for meta- -emotion by Gottman and collaborators, we tested 8 models of association between meta-emotion (separately for mothers and fathers), temperament (effortful control/negative affectivity) and children's problems (externalising/internalising). We chose to investigate children's negative affectivity and effortful con- trol as a possible link between mothers' and separately fa- thers' reports of family emotion variables and children's exter- nalising and internalising problems. Path analysis was used to evaluate our hypothesis that family emotion variables would be indirectly or/and directly related to a child's exter- nalising and internalising problems through or without neg- ative affectivity and effortful control. In other words, we set up models in which mothers' and fathers' awareness of their own and their child's emotions are both directly and indirect- ly associated with children's externalising and internalising problems, while mothers' and fathers' coaching of their child's emotions is only indirectly associated with children's prob- lems through the temperament dimension. Using path analy- sis we firstly estimated direct effects in all models, and after- wards indirect links were added to models and indirect ef- fects estimated. When some direct or indirect link was insi- gnificant, we omitted it from the models. Path analysis was used to derive the best-fitting model, which are presented in the figures. Analyses were performed using Mplus 7 (Muth?n & Muth?n, 1998-2012).

In view of mothers' ratings for awareness and coaching as well as effortful control as a dimension of temperament and their child's internalising problems, our model fits the data; ?2(1)= 0.735 (ns),CFI =1.00, TLI = 1.06, RMSEA = 0.00, SRMR = 0.01, and all paths were significant (Figure 1).

In this model, the R2 for effortful control was 0.027, and the R2 for internalising problems was 0.112. These findings indicate that mothers' awareness and coaching were related to child's effortful control and that child's effortful control was related to child internalising problems. The results for indi- rect relations (for awareness β = -0.03, p < 0.05 and coaching β = -0.04, p < 0.05) indicated that the indirect effect from mo- thers' awareness and coaching to child internalising problems through effortful control was significant. The model also con- firms the direct effect from mothers' awareness to child inter- nalising problems. The same model fits our data for externali- sing problems; χ2 (1) = 0.011 (ns), CFI = 1.00, TLI = 1.03, RMSEA = 0.00, SRMR = 0.001, and all paths were significant (Figure 2).

In this model, the R2 for effortful control and for externa- lising problems was 0.027 and 0.245 respectively, and for indi- rect relations it was for awareness β = -0.05, p < 0.05 and for coaching β = -0.06, p < 0.01. The same model fits our data for fa- thers' ratings for awareness and coaching (externalising pro- blems: χ2(1) = 1.006 (ns), CFI = 1.00, TLI = 1.00, RMSEA = 0.00, SRMR = 0.01; R2 for effortful control was 0.025, for externali- sing problems it was 0.244, for indirect relations it was for aware- ness β = -0.06, p < 0.01 and for coaching β = -0.05, p < 0.01; Fi- gure 2; internalising problems: χ2 (1) = 0.718 (ns), CFI = 1.00, TLI = 1.02, RMSEA = 0.00, SRMR = 0.01; R2 for effortful con- trol was 0.025, for internalising problems it was 0.124, for indi- rect relations for awareness β = -0.04, p < 0.05 and for coach- ing β = -0.03, p < 0.05; Figure 1). In summary, mothers' and fathers' awareness of their own and their child's emotions and mothers' and fathers' coaching of their child's emotions were indirectly related to their child's externalising and inter- nalising problems through their child's effortful control, but mothers' and fathers' awareness of their own and their child's emotion were also directly related to their child's externalis- ing and internalising problems.

The same model was found for mothers' ratings of aware- ness and coaching as well as negative affectivity as a dimen- sion of temperament and their child's externalising problems (χ2 (1) = 1.405 (ns), CFI = 1.00, TLI = 1.03, RMSEA = 0.00, SRMR = 0.01; R2 for negative affectivity was 0.056, for exter- nalising problems it was 0.134, for indirect relations for aware- ness β = -0.07, p < 0.01 and for coaching β = -0.03, p = 0.05; Figure 4). Another model was found for mothers' ratings of awareness and coaching as well as negative affectivity as di- mensions of temperament and child internalising problems (Figure 3).

Mothers' awareness of their own and their child's emotions and mothers' coaching of their child's emotions were indirect- ly related to child internalising problems through child nega- tive affectivity (χ2 (2) = 3.738 (ns), CFI = 0.98, TLI = 0.94, RMSEA = 0.04, SRMR = 0.02; for negative affectivity was 0.056, and for internalising problems was 0.085 and for indirect rela- tions for awareness β = -0.06, p < 0.01 and coaching β = -0.03, p = 0.05). The results seen in the last two models showed that only fathers' awareness of their own and their child's emo- tions was indirectly (through negative affectivity) and direct- ly related to child internalising (Figure 3) and externalising pro- blems (Figure 4). For child internalising problems: χ2 (3) = 0.790 (ns), CFI = 1.00, TLI = 1.07, RMSEA = 0.00, SRMR = 0.00; R2 for negative affectivity was 0.026, for internalising problems it was 0.104, and for indirect relations for awareness β = -0.04, p< 0.01. For child externalisingproblems: χ2(3) = 0.807 (ns), CFI =1.00, TLI =1.05, RMSEA = 0.00,SRMR=0.01; R2 for ne- gative affectivity was 0.026, for externalising problems it was 0.137 and for indirect relations for awareness β = -0.05, p < 0.01.

To conclude, the family emotion variables were indirect- ly related to child externalising and internalising problems through child negative affectivity and effortful control, and as demonstrated by our models, mothers' and fathers' aware- ness of their own and their child's emotions play a far more important role than coaching. Furthermore, mothers' and fa- thers' awareness of their own and their child's emotions were both directly (except for the model of mothers' meta-emotion, negative affectivity and internalising problems) and indirect- ly related to child externalising and internalising problems. The results of the models showed that mothers' and fathers' awareness and coaching were positively related to effortful control and negatively related to negative affectivity, and only mothers' and fathers' awareness was negatively related to externalising and internalising problems. Simultaneously, negative affectivity was positively related to child externalis- ing and internalising problems, while effortful control was negatively related to child externalising and internalising pro- blems. It is important to note that even though all the pro- posed models have been confirmed, the explained proportion of variances is relatively low. With these findings, we have confirmed part of the investigated assumptions from Got- tman's model relating to the influence of parental meta-emo- tion on child outcomes.

DISCUSSION

The results indicated that effortful control and negative affec- tivity partly mediated the relation between family emotion variables and child externalising and internalising problems. Effortful control was a more important variable in mediating the connection between family emotion variables and the child's problems. On the basis of these results obtained in our study, we can conclude that self-regulation plays a more im- portant role in the connection between parental meta-emo- tion and a child's problems. The results of the models showed that mothers' and fathers' awareness and coaching were posi- tively related to effortful control. The children of parents who are aware of their own and their children's emotions of anger and sadness and subsequently treat these children's emotions with respect, have a higher self-regulation capacity and vice versa. Moreover, children with higher effortful control have less externalising and internalising problems. Our models do not confirm such a clear relationship of associations between parental meta-emotion variables and the dimension of nega- tive affectivity. Greater mothers' and fathers' awareness of their own and children's emotions were associated with a less manifested dimension of children's negative affectivity and vice versa. Further, children with a higher negative affectivity have more externalising and internalising problems.

Furthermore, mothers' and fathers' awareness of their own and their child's emotions were directly (except for the model of mothers' meta-emotion, negative affectivity and internalis- ing problems) related to child externalising and internalising problems. Mothers and fathers with a greater degree of awareness of their own and their children's negative emo- tions have children with less problems; this relationship was indirectly associated through effortful control, i.e. negative affectivity but direct association was determined as well. In other words, lower levels of mothers' and fathers' awareness of their own and children's emotions were related to higher levels of internalising and externalising problems. It is impor- tant to note that this is applicable in complete functional fam- ilies because in families with intimate partner violence, fa- thers who were more aware of their children's fear had chil- dren who also showed higher levels of externalising problems than children whose fathers were less aware of their chil- dren's fear (Maliken & Katz, 2013). Thus, in the broader con- text of domestic violence, emotion coaching and child adjust- ment, domestic violence was associated with less coaching of anger and fear depending on whether the parent was the perpetrator or victim of domestic violence (Katz & Windeck- er-Nelson, 2006). Consequently, the results obtained in this study cannot be generalised and applied to various forms of dysfunctional families.

In addition, our results point to the importance of pa- rental meta-emotions in the development of preschool-aged children in Croatian families. The importance is manifold be- cause the development of a socially and emotionally compe- tent child in an adequate family environment in early child- hood ensures, and is a primary predisposing factor for psycho- social adjustment in adolescence. As confirmed by research, children with internalised and externalised problems, aggres- sive behaviour, a difficult temperament and with low self-e- steem, who develop in an unsatisfactory/inappropriate fami- ly environment (e.g. Denham et al., 2000; Eisenberg et al., 1995; Kochanska et al., 2000; Yap, Allen, Leve, & Katz, 2008) more frequently become psychosocially maladjusted later in childhood.

In view of parental emotion dimensions, our results do not support the relationship of Gottman's meta-emotion di- mension (Gottman et al., 1997; Katz & Windecker-Nelson, 2004), because they showed that parents who were higher on emotion coaching were not more aware of their child's emo- tions and vice versa. Moreover, in another investigation, no connection was determined between parents' awareness of their own and their child's emotions and coaching of child's emotions (Hooven et al., 1995), so awareness and coaching re- mained as two unconnected meta-emotion dimensions. It should be noted that we assumed that parents' coaching chil- dren's emotions in interaction with children's temperament characteristics, may influence children's psychosocial adjust- ment. In another study, mothers who tended to adopt an e- motion-coaching philosophy were more likely to achieve secure parent-child attachments (Chen, Lin, & Li, 2012). If we shift to the age of preadolescence, mother's acceptance of her own emotions was associated with fewer depressive symp- toms, higher self-esteem and fewer externalising problems in young adolescents (Katz & Hunter, 2007). The study by Chang, Schwartz, Dodge, and McBride-Chang, (2003) suggested that mothers' harsh parenting affected their child's emotion regu- lation more strongly than fathers', whereas harsh parenting emanating from fathers had a stronger effect on the child's aggression, and that negative emotionality was found as a potentially common cause of family perturbations, including parenting and child adjustment problems. It can be therefore concluded that family emotional life is a complex system which depends on attitudes and the conduct of mothers and fathers separately, but it also depends on a series of variables likely to affect a family as a system, such as parental styles, quality of marital relationship, stress in a family, etc. How- ever, on the basis of our results we could highlight the impor- tance of the role of maternal meta-emotion philosophy in cre- ating maternal child interaction in families and behavioural problems of children.

The possible limitations of our study undoubtedly include examination of meta-emotion dimensions. In our research, mothers and fathers completed the Meta-Emotion Structure Questionnaire. Similarly, as Hakim-Larson, Parker, Lee, Good- win, and Voelker (2006) in their research, we also started from a self-report version of the meta-emotion interview introduced in a parenting guidebook (Gottman, 1997). Lagacé-Séguin and Coplan (2005) developed a modified version of an existing self-assessment questionnaire designed to measure parents' e- motional style, and results from the study indicated a two- factor scale for the maternal emotional styles questionnaire, emotion coaching and emotion dismissing. On the other hand, Paterson and collaborators (2012) obtained different factorial structures of the parental meta-emotion dimension depend- ing on whether these were typically developing children or children with developmental disabilities (emotion coaching, parental acceptance of negative emotion, and parental rejec- tion of negative emotion and in subsample feelings of uncer- tainty/ineffectiveness in emotion socialisation). Therefore, it is hard to measure the construct of meta-emotion, partly be- cause it is both different and comparable to other constructs measuring family emotional life and dimensions of parent- hood critical for children's psychosocial adjustment. Based on the results of our and other studies, we can conclude that future studies should pay particular attention to measure- ment of the parental meta-emotion dimension, use different self-evaluation scales and combine interview and observation as an additional method. This should be done for separate dyads within family members as a system, separately for a subject sample from functional and dysfunctional families and depending on the child's development status (with de- velopmental disabilities or not).

To summarise, family emotion variables were indirectly related to child externalising and internalising problems through child negative affectivity and effortful control. In addition, mothers' and fathers' awareness of their own and their child's emotions were both directly and indirectly related to the child's externalising and internalising problems. The results of the models showed that mothers' and fathers' awareness and coaching were positively related to effortful control and negatively related to negative affectivity while only mothers' and fathers' awareness were negatively related to externalis- ing and internalising problems. Concurrently, negative affec- tivity was positively related to child externalising and inter- nalising problems, and effortful control was negatively rela- ted to child externalising and internalising problems. It is im- portant to note that even if all the proposed models are con- firmed, the explained proportion of variances are relatively low. With this study we have confirmed part of the investi- gated assumptions from Gottman's model relating to the in- fluence of parental meta-emotion on child outcomes.

These findings therefore point to a need for the develop- ment of an intervention programme aimed at improving pa- rental awareness and emotion coaching. Parent-child interac- tions help children to learn how to regulate their emotions which may increase their self-regulation ability, decrease ex- pressions of negative affectivity and result in fewer externali- sing and internalising problems. Parental education programs addressing parental attitudes and practices related to emo- tion should be supported, with special emphasis on the iden- tification of children at risk for emotion regulation difficulties. Preventive programmes are aimed at enhancing children's socio-emotional competence and taking steps to prevent the emergence of behavioural problems and psychopathology. Based on the analysis of a great deal of research, Izard (2002) suggests 7 principles for developing preventive interventions, which, in addition to emotion communication in early life, in- clude the utilisation of positive and negative emotions, emo- tion modulation as a mediator of emotion utilisation, emotion patterns in states and traits, different processes of emotion activation, and the development of connections for the mod- ular and relatively independent emotions and cognitive sys- tems. In association with the results obtained in this study, effective emotion communication in the child's early years can prevent emotional deprivation and dysfunctional emotion systems, foster the development of socio-emotional compe- tence and prevent behavioural problems such as aggression, violence, and social withdrawal.

Sidebar

Author's Note

The data presented in this paper were collected as a part of a comprehensive study on the emotional life of a family with preschool children.

References

REFERENCES

Achenbach, T. M. (1991). Manual for the Child Behavior Checklist/4-18 and 1991 Profile. Burlington, VT: University of Vermont Department of Psychiatry.

Ahadi, S. A., Rothbart, M. K., & Ye, R. M. (1993). Child temperament in the U.S. and China: Similarities and differences. European Journal of Personality, 7(5), 359-378. doi:10.1002/per.2410070506

Baumrind, D. (1971). Current patterns of parental authority. Develop- mental Psychology Monograph, 4(1, Pt. 2), 1-103. doi:10.1037/h0030372

Brajsa-Zganec, A. (2002). Parental meta-emotion structure questionnaire. Unpublished questionnaire.

Brajsa-Zganec, A. (2003). Dijete i obitelj: emocionalni i socijalni razvoj [Child and family - emotional and social development]. Jastrebarsko: Na- klada Slap.

Brajsa-Zganec, A., & Hanzec, I. (2014). Social development of preschool children in Croatia: Contributions of child temperament, maternal life satisfaction and rearing practices. Journal of Child and Family Studies, 23(1), 105-117. doi:10.1007/s10826-012-9696-8

Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulle- tin, 56(2), 81-105. doi:10.1037/h0046016

Chang, L., Schwartz, D., Dodge, K. A., & McBride-Chang, C. (2003). Harsh parenting in relation to child emotion regulation and aggression. JournalofFamily Psychology, 17(4), 598-606.doi:10.1037/0893-3200.17.4.598

Chen, F. M., Lin, H. S., & Li, C. H. (2012). The role of emotion in par- ent-child relationships: Children's emotionality, maternal meta-e- motion, and children's attachment security. Journal of Child and Fa- mily Studies, 21(3), 403-410. doi:10.1007/s10826-011-9491-y

Cowan, P. A. (1996). Meta-thoughts on the role of meta-emotion in children's development: Comment on Gottman et al. (1996). Journal of Family Psychology, 10(3), 277-283. doi:10.1037/0893-3200.10.3.277

Darling, N., & Steinberg, L. (1993). Parenting style as context: An in- tegrative model. Psychological Bulletin, 113(3), 487-496. doi:10.1037/ 0033-2909.113.3.487

Denham, S. A. (1997). "When I have a bad dream, mommy holds me": Preschoolers' conceptions of emotions, parental socialisation, and emotional competence. International Journal of Behavioral Development, 20(2), 301-319. doi:10.1080/016502597385351

Denham, S. A., Workman, E., Cole, P. M., Weissbrod, C., Kendziora, K. T., & Zahn-Waxler, C. (2000). Prediction of externalizing behavior problems from early to middle childhood: The role of parental so- cialization and emotion expression. Development and Psychopathology, 12(1), 23-45. doi:10.1017/S0954579400001024

Dettling, A. C., Gunnar, M. R., & Donzella, B. (1999). Cortisol levels of young children in full-day childcare centers: Relations with age and temperament. Psychoneuroendocrinology, 24(5), 519-536. doi:10.1016/ S0306-4530(99)00009-8

Eisenberg, N. (1996). Meta-emotion and socialization of emotion in the family - A topic whose time has come: Comment on Gottman et al. (1996). Journal of Family Psychology, 10(3), 269-276. doi:10.1037/0893- 3200.10.3.269

Eisenberg, N., & Fabes, R. A. (1995). The relation of young children's vicarious emotional responding to social competence, regulation and emotionality. Cognition and Emotion, 9(2-3), 203-228. doi:10.1080/026 99939508409009

Eisenberg, N., Cumberland, A., Spinrad, T. L., Fabes, R. A., Shepard, S. A. et al. (2001). The relations of regulation and emotionality to chil- dren's externalizing and internalizing problem behavior. Child De- velopment, 72(4), 1112-1134. doi:10.1111/1467-8624.00337

Eisenberg, N., Fabes, R. A., & Losoya, S. (1997). Emotional respond- ing: Regulation, social correlates, and socialization. In P. Salovey, & D. J. Sluyter (Eds.), Emotional development and emotional intelligence: Edu- cational implications (pp. 129-163). New York: Basic Books.

Eisenberg, N., Fabes, R. A., Murphy, B., Maszk, P., Smith, M., & Kar- bon, M. (1995). The role of emotionality and regulation in children's social functioning: A longitudinal study. Child Development, 66(5), 1360-1384. doi:10.2307/1131652

Goldsmith, H. H., & Rothbart, M. K. (1991). Contemporary instru- ments for assessing early temperament by questionnaire and in the laboratory. In J. Strelau, & A. Angleitner (Eds.), Explorations in tempera- ment (pp. 249-274). London: Plenum Press. doi:10.1007/978-1-4899- 0643-4_16

Gottman, J. M. (1997). Kinder Brauchen Emotionale Intelligenz. Muen- chen: Diana Verlag.

Gottman, J. M., Katz, F. L., & Hooven, C. (1996). Parental meta-emo- tion philosophy and the emotional life of families: Theoretical models and preliminary data. Journal of Family Psychology, 10(3), 243-268. doi:10. 1037/0893-3200.10.3.243

Gottman, J. M., Katz, F. L., & Hooven, C. (1997). Meta-emotion: How fami- lies communicate emotionally. Mahwah: Lawrence Erlbaum Associates.

Hakim-Larson, J., Parker, A., Lee, C., Goodwin, J., & Voelker, S. (2006). Measuring parental meta-emotion: Psychometric properties of the emotion-related parenting styles self-test. Early Education and Deve- lopment, 17(2), 229-251. doi:10.1207/s15566935eed1702_2

Hooven, C., Gottman, J. M., & Fainsilber Katz, L. (1995). Parental me- ta-emotion structure predicts family and child outcomes. Cognition and Emotion, 9(2-3), 229-264. doi:10.1080/02699939508409010

Hunter, E. C., Katz, L. F., Wu Shortt, J., Davis, B., Leve, C., Allen, N. B., & Sheeber, L. B. (2011). How do i feel about feelings? Emotion socialization in families of depressed and healthy adolescents. Jour- nal of Youth Adolescence, 40(4), 428-441. doi:10.1007/s10964-010-9545-2

Izard, C. E. (2002). Translating emotion theory and research into pre- ventive interventions. Psychological Bulletin, 128(5), 796-824. doi:10. 1037/0033-2909.128.5.796

Katz, L. F., & Gottman, J. M. (1986). The meta-emotion interview. Un- published manual. Seattle: University of Washington.

Katz, L. F., & Gottman, J. M. (1993). Patterns of marital conflict pre- dict children's internalizing and externalizing behaviors. Develop- mental Psychology, 29(6), 940-950. doi:10.1037/0012-1649.29.6.940

Katz, L. F., & Gottman, J. M. (1997). Buffering children from marital con- flict and dissolution. Journal of Clinical Child Psychology, 26(2), 157-171. doi:10.1207/s15374424jccp2602_4

Katz, L. F., Gottman, J. M., & Hooven, C. (1996). Meta-emotion philo- sophy and family functioning: Reply to Cowan (1996) and Eisenberg (1996). Journal of Family Psychology, 10(3), 284-291. doi:10.1037/0893- 3200.10.3.284

Katz, L. F., & Windecker-Nelson, B. (2004). Parental meta-emotion philosophy in families with conduct-problem children: Links with peer relations. Journal of Abnormal Child Psychology, 32(4), 385-398.

Katz, L. F., & Hunter, E. C. (2007), Maternal meta-emotion philoso- phy and adolescent depressive symptomatology. Social Development, 16(2), 343-360. doi:10.1111/j.1467-9507.2007.00388.x

Katz, L. F., & Windecker-Nelson, B. (2006). Domestic violence, emo- tion coaching, and child adjustment. Journal of Family Psychology, 20(1), 56-67. doi:10.1037/0893-3200.20.1.56

Katz, L. F., Maliken, A. C., & Stettler, N. M. (2012). Parental meta-emo- tion philosophy: A review of research and theoretical framework. Child Development Perspectives, 6(4), 417-422. doi:10.1111/j.1750-8606. 2012.00244.x

Kochanska, G., Murray, K., & Coy, K. C. (1997). Inhibitory control as a contributor to conscience in childhood: From toddler to early school age. Child Development, 68(2), 263-277. doi:10.2307/1131849

Kochanska, G., Murray, T. K., & Harlan, E. (2000). Effortful control in early childhood: Continuity and change, antecedents, and implications for social development. Developmental Psychology, 36(2), 220-232. doi:10.1037/0012-1649.36.2.220

Kochanska, G., Murray, K., Jacques, T. Y., Koenig, A. L., & Vandegeest, K. A. (1996). Inhibitory control in young children and its role in emer- ging internalization. Child Development, 67(2), 490-507. doi:10.2307/ 1131828

Lagacé-Séguin, D. G., & Coplan, R. J. (2005). Maternal emotional styles and child social adjustment: Assessment, correlates, outcomes and goodness of fit in early childhood. Social Development, 14(4), 613-636. doi:10.1111/j.1467-9507.2005.00320.x

Lansford, J. E., Wager, L. B., Bates, J. E., Dodge, K. A., & Pettit, G. S. (2012). Parental reasoning, denying, privileges, yelling and spank- ing: Ethnic differences and associations with child externalizing beha- vior. Parenting Science and Practice, 12(1), 42-56. doi:10.1080/15295192. 2011.613727

Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: Parent-child interaction. In P. H. Mussen (Ed.), Handbook of child psychology: Socialization, personality, and social development (pp. 1-101). New York: Wiley.

Maliken, A. C., & Katz, L. F. (2013). Fathers' emotional awareness and children's empathy and externalizing problems. Journal of Interperso- nal Violence, 28(4), 718-734. doi:10.1177/0886260512455866

Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey, & D. J. Sluyter (Eds.), Emotional development and emotional intelligence: Educational implications (pp. 3-31). New York: Basic Books.

Muthén, L. K., & Muthén, B. O. (1998-2012), Mplus User's Guide. Se- venth Edition. Los Angeles, California: Muthén and Muthén.

Paterson, A. D., Babb, K. A., Camodeca, A., Goodwin, J., Hakim-Lar- sen, J., Voelker, S., & Gragg, M. (2012). Emotion-Related Parenting Styles (ERPS): A short form for measuring parental meta-emotion philosophy. Early Education and Development, 23(4), 583-602. doi:10. 1080/10409289.2011.569316

Ramsden, S. R., & Hubbard, J. A. (2002). Family expressiveness and parental emotion coaching: Their role in children's emotion regulation and aggression. Journal of Abnormal Child Psychology, 30(6), 657-667. doi:10.1023/A:1020819915881

Rothbart, M. K., Ahadi, S. A., & Hershey, K. L. (1994). Temperament and social behavior in childhood. Merrill-Palmer Quarterly, 40(1), 21-39.

Russell, A., Hart, C. H., Robinson, C. C., & Olsen, S. F. (2003). Chil- dren's sociable and aggressive behaviour with peers: A comparison of the US and Australia, and contributions of temperament and par- enting styles. International Journal of Behavioral Development, 27(1), 74-86. doi:10.1080/01650250244000038

Stright, A. D., Gallagher, K. C., & Kelley, K. (2008). Infant tempera- ment moderates relations between maternal parenting in early childhood and children's adjustment in first grade. Child Develop- ment, 79(1), 186-200. doi:10.1111/j.1467-8624.2007.01119.x

Van Zeijl, J., Mesman, J., Stolk, M. N., Alink, L. R. A., Van IJzendoorn, M. H., Bakermans-Kranenburg, M. J., Juffer, F., & Koot, H. M. (2007). Differential susceptibility to discipline: The moderating effect of child temperament on the association between maternal discipline and early childhood externalizing problems. Journal of Family Psychology, 21(4), 626-636. doi:10.1037/0893-3200.21.4.626

Wu Shortt, J., Stoolmiller, M., Smith-Shine, J. N., Eddy, J. M., & Shee- ber, L. (2010). Maternal emotion coaching, adolescent anger, regulation, and siblings' externalizing symptom. Journal of Child Psychology and Psychiatry, 51(7), 799-808.

Yap, M. B. H., Allen, N. B., Leve, C., & Katz, L. F. (2008). Maternal meta-emotion philosophy and socialization of adolescent affect: The moderating role of adolescent temperament. Journal of Family Psycho- logy, 22(5), 688-700. doi:10.1037/a0013104