Parental sense of competence, resilience, and empathy in relation fathers’ responses to children’s negative emotions in the context of everyday paternal childrearing decisions

Abstract It is expected that children are going to respond with anger and resentment to some of their parents’ child rearing and disciplinary decisions. Parents may perceive intense emotional reactions by children as an “attack” on them, challenging their ability to provide a supportive parental thoughtful responses. The present study examines the role of parental sense of competence and parental resilience, as well as parental empathy and aggression in fathers’ reactions to children’s negative emotions (FRCNE) in the context of such parental decisions. Study 1, assessed the role of paternal competence and resilience in fathers’ reactions to their children’s negative emotions. Findings showed that both paternal competence and resilience were associated with less negative parental reaction such as punishment toward the child but not with more positive reaction like calming down the child. Extending Study 1, Study 2, examined whether paternal empathy and aggression would associate with paternal positive responses. Findings showed that paternal greater empathy was associated with fathers’ higher likelihood to respond positively to child frustrations. Considered together, the findings of the two studies suggest that the importance of both paternal resilience and empathy in addressing and managing children’s reactions to frustration. Implications for intervention are discussed.

ABOUT THE AUTHORS Dr. Ido Ziv is a Lecture in the Department of Psychology, the College of Management Academic Studies, Israel. He serves as a mutual head of the Graduate Program in Adolescent and Young Adult Clinical Psychology. He is a Clinical Psychologist and is a member of the International Association for Relational psychoanalysis (IRAPP) and the International Association for Psychoanalytic Self Psychology (IAPSP). His Research Interests are adolescents and young adults' romantic relations, parental presence, skills and practices. As well he has a great interest in developing and using new technologies for symptoms alleviation and as a complementary tool for diagnosis and therapy.
Ms. Nadia Golbez works as a Psychologist. She owns a MA degree in clinical psychology.
Mrs. Nili Shapira works as a Psychologist. She owns a MA degree in social psychology.

PUBLIC INTEREST STATEMENT
Parenting, writ large, refers to parents' ability to exert discipline while meeting their children's needs for warmth and love, for cognitive, emotional and behavioral support, and for help developing their emotion regulation capacities. During the day-to-day process of raising children, parents must often make decisions (e.g., setting bedtimes) which contradict their child's wishes and may trigger child negative reactions. Parents must then manage the task of regulating their own emotions in order to provide the appropriate mix of discipline and emotional support to the child. The current study asks what are the paternal practices that are related to fathers positive (e.g., calming down the child) or negative (e.g., punishment) response at these emotionally intense situations? Findings show that paternal competence and resilience were associated with less negative parental reaction and that paternal greater empathy was associated with fathers' higher likelihood to respond positively to child frustrations. Implications for intervention are discussed.

Introduction
A large body of research in psychology deals with parental reactions 1 to children's negative emotions (PRCNE)-that is, how parents deal with children's everyday frustrations and distress (e.g., crying over a broken toy). Despite this vast literature, however, little is known about the internal factors that enable parents to provide supportive parenting (Goeke-Morey & Cummings, 2007;Meyer et al., 2014;Valiente et al., 2007). This is particularly true for fathers, whose role in parenting is often still not fully acknowledged (Fitzgerald & Bradley, 2012;McElwain et al., 2007;Nelson et al., 2009;Stolz et al., 2005). Though the role of fathers in children's lives is slowly gaining more research attention. Within this scope it is important also to learn, how fathers' own attributes affect their reactions (Castillo et al., 2011).
The present research aims to help bridge this gap by examining the relation between positive and negative paternal reactions and two paternal attributes: A. general attributes: resilience and self-efficacy (i.e., parental sense of competence). B. emotional attributes: empathy and aggression. In highlighting resilience and competence, we follow Bandura's self-efficacy model (Bandura, 1997;Bandura et al., 2011), which suggests that these attributes are foundational to successful coping in the face of obstacles. We chose empathy and aggression-two opposing emotional ways of relating to others-because of the former's well-established relationship with supportive parenting (Manczak et al., 2016). The proposed conceptual model suggests that in the face of child frustration associated with paternal child rearing daily decisions. paternal resilience and competence are essential to reduce hostile parenting reactions toward the child, while paternal emotional attributes will increase positive parenting responses (see Figure 1). The first part of the conceptual model is examined in Study 1 while the latter in Study 2.

Parental reactions to children's negative emotions
Since the nature of negative emotions is aversive, parents are often motivated to react to them by using negative control strategies (Aunola et al., 2013). These strategies-from overt hostility and punishment to withdrawal-are all associated with maladaptive adjustment on the part of children. For instance, punishment has been found to increase tensions in the parent-child relationship (Power, 2004) and externalizing behavior by the child (Tao et al., 2010). A reaction that minimizes or devalues the child's feelings is associated with lower social competence (Fabes et al., 2001;Eisenberg et al., 1998), while hostility and withdrawal have been linked with scholastic maladjustment (Sturge-Apple et al., 2006). In contrast, supportive parental responses-whether emotion-focused, such as comforting the child or helping the child work through his feelings, or problem-focused, such as helping the child think of ways to resolve the issue of concern-enhance children's psychological adjustment (Fabes et al., 2002). More specifically, these responses strengthen positive parent-child relationships and thereby improve children's emotional security (Cassidy, 1994). In addition, these strategies help children internalize their parents' developed emotional abilities, thereby improving their own emotion management and regulation (Gottman et al., 1997;Power, 2004). Problem-focused and emotionfocused responses have been positively related to children's social skills (Eisenberg et al., 1996(Eisenberg et al., , 1999Gottman et al., 1997), and negatively related to internalizing problems (Tao et al., 2010).
Besides the general distinction between supportive and unsupportive parenting, scholars distinguish between different types of unsupportive reactions, and in particular, between what we may call negative and detached reactions. Negative parental reactions include those with a punitive component, whether via outright punishment (e.g., sending the child to his room) or through expressions of hostility and anger. Another negative parental reaction is minimization, in which the parent plays down the importance of the situation or devalues the child's feelings. Detached reactions come in two forms. The first is withdrawal, where the parent makes himself emotionally and even physically unavailable (e.g., by ignoring the child or leaving the room). Sturge-Apple et al. (2006) found that the withdrawal reactions, even more than parental hostility, was directly associated with subsequent increases in children's psychological problems and scholastic maladjustment. The second detached reaction is giving in-i.e., letting the child do what he wants while feeling frustrated and angry.

Parental resilience and sense of competence
Parenting, writ large, refers to parents' ability to exert authority and discipline while meeting their children's needs for warmth and love, for cognitive, emotional and behavioral support, and for help developing their emotion regulation capacities (Baumrind, 2005). During the day-to-day process of raising children, parents must often make decisions (e.g., setting bedtimes) which contradict their child's wishes and may trigger child negative reactions. Parents must then manage the task of regulating their own emotions in order to provide the appropriate mix of discipline and emotional support to the child.
Research into the family lives of children suggests that parents who feel more confident in their parenting abilities tend to be more effective at managing their own feelings and behaviors in order to support the child. While this seems logical (those who are better at raising children are likely to feel more confident in their ability to do so), evidence suggests that the causal relationship reflected here is not one-directional. Rather, it has been found that when parents feel confident in their ability to parent, they are likely to use more effective parenting practices which foster positive emotional development outcomes for their child (Gilmore & Cuskelly, 2008). Contrarily, in studies of parents whose children were coping with hyperactivity, impulsivity and attention problems, parents' feelings of powerless were associated with an increase in negative parenting behaviors over time (Glatz & Stattin, 2013;Glatz et al., 2011). Parents were also found to react in less supportive ways to anxious children (Hurrell et al., 2015). Similarly, Grossklaus and Marvicsin (2014), in a detailed review of the literature on parenting efficacy and childhood obesity, point to parental satisfaction and parental selfefficacy as contributing factors to children's eating habits. These findings make sense in light of Bandura self-efficacy model (Bandura, 1997).under which parents with a low sense of power perceive their children's problematic behaviors as difficult to change, and these perceptions then make parents behave toward their children in negative and unsupportive ways (Bugental, 2009).
Even in families not coping with ADHD or other disorders, children's negative emotional displays (e.g., of anger, resentment or distress) may be associated with parents frustration and powerless. In the case of older children whose parent (in our study, the father) has made a decision that angers or upsets the child, the father is likely to perceive the child's reaction as a hostile attack-a situation that may be emotionally difficult for the father to handle. We expect that a strong belief on the part of the father that he can cope with such a situation (i.e., parental sense of competence) will indeed help him to do so. In practice, this means that fathers with a strong sense of their own competence in the parental role will respond to the child in a healthy and supportive way, whereas fathers lacking such a sense of competence will react with negative control strategies, or alternatively will remove themselves from the situation through detachment or withdrawal.
One factor that may contribute to a sense of competence in parents, and allow them to cope with problematic behaviors or negative emotions in their children, is resilience (Gibaud-Wallston & Wandersman, 1978;Gilmore & Cuskelly, 2008;Ohan, Leung, & Johnston, 2000). Past research has defined resilience as the ability to "bounce back" from challenges, including internal and external stressors (Windle et al., 2011). While other scholars see resilience as a more developing attribute as a result of coping with stressful situations (Henry et al., 2015). The current definition is focused on resilience as a coping and protecting mechanism through available resources.
Resilience has been linked to five important attributes: (1) a sense of personal competence; (2) trust in one's instincts; (3) positive acceptance of change; (4) a sense of control; and (5) spiritual influence (Connor & Davidson, 2003). Friborg et al. (2005) found the resilience construct to consist of five factors: personal competence, social competence, personal structure, family coherence, and social support.
These findings suggest that what is called resilience actually comprises a mix of several attributes or abilities that contribute to an individual one dimension capacity to cope with challenges and stressors. Against this backdrop, resilience is likely to play a crucial role when it comes to dealing with children's negative emotions, especially when these emotions are engendered by the father's decisions and can therefore be perceived by the father as an attack on himself. In such situations, resilience may support fathers' sense of competence as parents, and help them display the confidence and calmness required for supportive parenting.

Fatherhood and FRCNE
Despite growing acknowledgement of fathers' contributions to their children's emotional development (Cabrera et al., 2014;Fitzgerald et al., 2015) and educational outcomes (Jeynes, 2015;Lewis & Sussman, 2013;Sturge-Apple et al., 2006), there are only few studies of fathers reactions to children' negative emotions. Studies which included fathers have found that they tend to react with unsupportive reactions, such as minimization, anger, or punishment (Cassano et al., 2007;Eisenberg et al., 1996). What is more, paternal withdrawal and emotional unavailability, even more so than maternal displays of the same behaviors, were found to be associated with children's psychological problems and external symptoms such as reduced social competence and school adjustment (Sturge-Apple et al., 2006). Other studies have also found a particular impact of fathers' maladaptive reactions to children's emotional dysregulation on the development of psychopathology in both boys and girls (Suveg et al., 2008;Thomassin & Suveg, 2014).
Given the overall paucity of studies addressing FRCNE, such findings raise the question of whether fathers' reactions to their children's emotions tend to be similar across the board, or whether they might be influenced by intervening factors. One such potential intervening factor is the intensity of the child's emotional display. Fabes et al. (2001) found that children's emotional intensity mediated the relationship between parental coping reactions and children's social competence, such that harsh parental reactions to children's negative emotions were associated with both greater child emotional intensity and lower social competence in children. Thus, we argue that fathers' reactions in these situations will be influenced by the intensity of the child's emotional reaction.

FRCNE and gender
Another question that arises is whether fathers are likely to respond differently to negative emotions in daughters and sons. Lytton and Romney (1991), in an important meta-analysis, suggest that, overall, parents do not systematically differ in their childrearing practices for boys and girls, though a few differences were found. Specifically, in North American studies, both parents encouraged sex-typed activities, and in other Western nations, physical punishment was applied significantly more to boys. In addition, of relevance to the present study, fathers tended to differentiate more than mothers between boys and girls (Lytton & Romney, 1991).
More recent studies have found some support for parental differentiation by the child's gender. For instance, Cabrera et al. (2012) found that when the family space was very chaotic, boys were more often physically punished (spanked) by fathers and were less supported by mothers (these trends were not seen when the environment was less chaotic). Perry et al. (2015), in research based on adults' childhood memories, found that mothers were slightly more likely to react punitively to negative emotions in their sons than in their daughters, and mothers' punitive reactions toward their daughters were moderated by emotional closeness, an effect that was not found for boys. Other recent studies have found that parents, and particularly mothers, discuss emotions with their daughters more than with their sons, and are more likely to discourage anger and aggression in their daughters (Chaplin et al., 2005;Klimes-Dougan & Zeman, 2007). Parents also appear to use a greater number and variety of emotion terms with daughters than with sons (Adams et al., 1995;Kuebli et al., 1995). On the other hand, the literature is not homogeneous. For instance, Hurrell et al. (2015) found no significant differences between girls and boys for maternal or paternal supportive and non-supportive parenting.
On the whole, extant research suggests some support for the claim that parents, including fathers, treat emotional connections as particularly important for daughters, on the basis that women are thought to place more value on interpersonal relationships than men (Ryan et al., 2005). We therefore expect to find an interaction between the child's gender and FRCNE.

The present studies
As explored above we proposed a conceptual model to focus on the determinants of fathers' reactions to negative emotions in children, rather than their consequences. The model suggests that in the face of child frustration paternal resilience and competence are essential to reduce hostile parenting reactions toward the child, while paternal emotional attributes will increase parenting positive responses. Study 1 concentrated on the first part of the model. On the basis of the above summary we hypothesized for study 1 the following: Hypothesis 1: Parental sense of competence will be negatively associated with unsupportive (i.e., negative and detached) FRCNE and positively associated with supportive FRCNE (see Figure 2).
Hypothesis 2: Resilience will be negatively associated with unsupportive (negative or detached) FRCNE and positively associated with supportive FRCNE (see Figure 2).
Hypothesis 3: Resilience and parental sense of competence will be positively and strongly related.
Hypothesis 4: Resilience will be indirect associated with FRCNE through Parental sense of competence. Hypothesis 5: Increasing intensity of the child's emotional response will be associated with a decline in supportive FRCNE and an increase in unsupportive (negative and detached) FRCNE.
Hypothesis 6: Fathers will display a more supportive paternal attitude toward girls, and more negative FRCNE toward boys.
We examined the study questions and hypotheses in an Israeli sample. Israel is a developed, industrialized and Western culture that emphasizes individualistic values (Schwartz, 1994). An important characteristic of Israeli society, reflecting a moderately high level of traditional values, is the high importance ascribed to the family and community (Lavee & Katz, 2003;Scharf & Mayseless, 2010). Israeli culture is highly child-oriented (Lavee & Katz, 2003), and following one's children growing up is among the greatest joys of life. Conversely, children retain close and frequent contact with family members, and especially with parents, throughout their lives (Scharf, 2014).

Participants and procedure
Two hundred and thirty-six participants, all fathers of adolescents between 11 and 15 years of age, participated in an online survey in exchange for approximately 10. USD Participants were recruited from members of a well-known online polling service via a link sent by email. The link remained open until the target number of participants completed the questionnaire, a process that took about two days. The 236 participants in the final sample comprised all those who completed the full survey (a condition met by most respondents who opened the link) and who did not meet the exclusion criteria described below. IRB approval was obtained at the start of the process.
Participants were asked to relate in their answers only to children currently in the 11-15 age group; if they had more than one child in that age group, they were asked to relate to the child who was closest to 13 years of age. The sample excluded fathers whose children had any known diagnosed disorder-e.g., ADHD, anxiety stress disorder, or psychosis-that might put greater demands on the father's parenting skills. Participants who did not share the same residence as their children and participants who defined themselves as orthodox 2 were also excluded.
Participants mean age was 45.70 years (SD = 5.73). 233 were married and 3 divorced. 95% of them lives in cities (pop.>250,000) while the rest in towns (pop.<250,000). Participants earned average wages in Israel. The mean number of children in the family was 2.33 (SD =.62). Participants (beside four) had completed their high school education M = 14.89 years (SD = 2.57). 49% of the participants related to their boys while 51% to their girls.

Measures
Fathers' reactions to children's negative emotions were assessed using an instrument adapted from Fabes et al. 's (1990) Coping with Children's Negative Emotions Scale (CCNES). The CCNES lists 12 everyday situations that can lead young children to exhibit distress (e.g., losing a prized possession; being rejected by other children in the playground). Each situation is followed by six possible parental responses reflecting various supportive and unsupportive reactions. For each situation, respondents are asked to indicate how likely (on a scale from 1 = "very unlikely" to 7 = "very likely") they are to respond in each of the suggested ways. Evaluation of the CCNES has revealed good internal consistency, and high test-retest reliability over both a four-month (Fabes et al., 2002) and six-year period (Eisenberg et al., 1999).
For the present study, the CCNES was modified in four ways. First, the basic situations were altered to suit school-age children (e.g., not being allowed to stay overnight at a friend's house; being set a curfew). Second, the situations were extended so as to include a decision, demand or action by the father, followed by the child's reaction (e.g., "Your child records a movie, and you accidentally delete it. He gets angry and stamps his feet"). Third, following Fabes et al. (2001), the situations were designed so as to reflect three levels of children's emotional intensity: (1) low (a non-aggressive verbal display; e.g., "That's not fair!"); (2) moderate (an aggressive verbal display; e.g., "You're disgusting!"); and (3) high (an aggressive verbal and physical display). The number of situations was increased from 12 to 18 to allow six situations for each level.
The fourth modification related to the list of parental reactions, which were extended from six to eight. The six original reactions were designed to capture the following parental response strategies: (a) problem-focused reactions (helping the child solve the problem); (b) emotion-focused reactionsabreaction (helping the child express his/her feelings); (c) emotion-focused reactions-comfort; (d) punitive reactions-aversive (punishment); (e) punitive reactions-hostile (expressing anger at the child); and (f) minimization (devaluing the child's feelings). Two strategies were added for the current study: (g) parental withdrawal (making oneself emotionally unavailable; Sturge-Apple et al., 2006); and (h) giving in (i.e., letting the child do what he wants). In line with Fabes et al. (2002) and supporting research (e.g., Hurrell et al., 2015;McElwain et al., 2007;Suveg et al., 2008), responses a, b and c are considered to be positive or supportive, while responses d, e and f are classified as negative and responses g and h as detached. As in the original CCNES scale, respondents were asked to evaluate on a 7-point scale how likely they were to follow the course of action described, where 1 = "very unlikely" and 7 = "very likely". The full questionnaire can be found in Appendix 1.
Resilience was measured using Friborg et al. 's (2005) Resilience Scale for Adults, which Windle et al. (2011), in a meta-analysis comparing 19 resilience measures, found to receive among the best psychometric ratings. This scale comprises 33 items, each of which presents subjects with a situation (e.g., "When something unforeseen happens") followed by a five-point scale anchored by two routes of action: positive (e.g., "I always find a solution") and negative (e.g., "I often feel bewildered"). Subjects are asked to mark their reactions on the scale. To reduce bias, the positive possibility is presented on the left side of the scale in half the items and on the right in the other half. The questionnaire reflects five factors (see also Hjemdal et al., 2001), such that three factors measure various aspects of personal competence (present and future personal strength, 5 items each; social competence, 6 items; self-structured style, 4 items), while two factors measure individual sources of support (family cohesion, 6 items; social resources, 7 items). The scale yields a Cronbach's alpha of .88. The subscales yield varied moderate reliabilities (Personal strength-Current perception of the self α = 0.73; Personal strength-Future perception of the self α = 0.71; Social competence α = 0.65; Structured style α = 0.79; Family cohesion α = 0.78; Social resources α = 0.78).
Sense of Competence was measured using the Parenting Sense of Competence Scale (PSOC) of Gibaud-Wallston and Wandersman (1978, as cited in Johnston & Mash, 1989). The PSOC is the most commonly used tool for measuring parents' satisfaction with parenting (e.g., "A difficult problem in being a parent is not knowing whether you're doing a good job or a bad one") and self-efficacy in their parenting role (e.g., "The problems of taking care of a child are easy to solve once you know how your actions affect your child, an understanding I have acquired") (Jones & Prinz, 2005). Parents indicate their agreement with each item by circling a number between 1 (strongly agree) and 6 (strongly disagree). Several items are reverse-scored so that high scores indicate a positive parental experience. Suitable internal consistency (in the range 0.75-0.88) has been reported for the PSOC in a number of studies (Johnston & Mash, 1989;Lovejoy et al., 1997;Ohan, Leung, & Johnston, 2000).
Originally the scale contained 17 items. The version used in the current study, following Rogers and Matthews (2004) and Gilmore and Cuskelly (2008), included 14 items covering three factors: parental satisfaction (6 items), parental self-efficacy (5 items) and parental interest (3 items).

Results
The analysis is presented in three parts. First, we present the descriptive analyses, which address the first three hypotheses (H1-H3). Second, we provide the results for the indirect association hypothesis (H4). Finally, we examine whether differences in paternal reactions to children's negative emotions are related to the intensity of children's reactions (H5) and to the child's gender (H6). For this purpose, we used analysis of variance (ANOVA), based on a split-plot design with one between-subjects factor and two within-subject factors. Table 1 presents the descriptive analyses-i.e., means, standard deviations, and intercorrelations between the dependent variable (FRCNE) and the independent variables (resilience and PSOC)-for the three hypothesized models (supportive FRCNE, negative FRCNE and detached FRCNE). The intercorrelations partially support our first three hypotheses. First, paternal sense of competence was negatively correlated with negative and detached FRCNE, as posited by H1, though there was no positive correlation with supportive FRCNE. This trend held for all three levels of emotional intensity. Second, overall, the resilience scale behaved like the sense of competence scale: it was negatively correlated with detached and negative FRCNE for the three intensity levels, though to a lesser degree, and it was also not positively correlated with supportive FRCNE (H2). Finally, resilience was positively correlated with paternal sense of competence (H3).

Indirect association between resilience and FRCNE through PSOC
As shown in Table 1, there was no correlation between resilience and supportive FRCNE (model 1). Therefore, we conducted analysis only for model 2 and model 3, to test whether the negative indirect association between resilience and negative or detached FRCNE was through sense of competence. Also, since the correlation analyses just described revealed similar patterns for the three intensity levels, before executing the indirect analysis we averaged the results for the three levels for each parental response tendency. We therefore present two indirect association regression results, for model 2 and model 3, rather than six (two types of responses X three levels of intensity).
In our analyses we followed the recommendations of Preacher and Hayes (2008) and Hayes (2009), and we used the SPSS macro developed by Hayes and Preacher (2014). In the first indirect association analysis, we examine the relation between resilience and negative FRCNE through PSOC (Model 2). In the second indirect association analysis, the relation between resilience and detached FRCNE through PSOC was examined (Model 3). For both analyses, we tested the effect of resilience on FRCNE (total association, path c'+ab) and the effect of resilience on FRCNE through PSOC (direct association, path c'). Bootstrapping with 5,000 samples was used to formally test whether the strength of the difference between the total association (path c'+ab) and the direct association (path c')-the indirect association-was significantly different from zero (Preacher & Hayes, 2008). When the confidence intervals (CIs) of bootstrapping point estimates do not contain zero, the indirect association is considered statistically significant. We also computed the ratio of the indirect association to the total association as recommended by Preacher and Kelly (2011). Before conducting the analysis all variables were standardized.
For both indirect association analyses resilience was significantly associated with the predicted variable (negative FRCNE and detached FRCNE) and with PSOC, and PSOC was significantly associated with the outcome variable. Further, the direct association between resilience and the two forms of FRCNE were no longer significant after PSOC was included. The results are shown in   association was −.41, 95% CI (−0.61, −0.23) for negative FRCNE and −.53, 95% CI (−0.71, −0.38) for detached FRCNE, accounting for 100% and 88% of the total association, respectively. These findings indicate that regardless of whether the paternal reaction is negative or detached, the (negative) association between resilience and the paternal reaction is indirect through paternal sense of competence. Hypothesis 4 is therefore supported for models 2 and 3.

Childs's gender and emotional intensity as moderators of FRCNE
We now turn to our final hypotheses for Study 1-namely, the expectation that paternal reactions to children's negative emotions will differ according to the intensity of the child's emotions (H5) and/or the child's gender (H6). For this purpose, we used a mixed-design 8 × 3 X 2 ANOVA (father's reaction X child's intensity X child's gender). We treated the father's reactions and child's intensity as within-participant variables, and the child's gender as a between-participants variable.
Before conducting the analysis, we ensured that there were no outliers in the data (2 SD above and below the mean; see Ratcliff, 1993). Furthermore, prior analysis demonstrated that skewness was < 1 for all variables and that all criteria relating to kurtosis were satisfactory. The main effect for FRCNE was significant, F(7,1652) = 252.8, p < .001, ηp2 = 0.51. Also, though we did not hypothesize regarding the relative weight of the supportive versus non-supportive reaction types, it was noteworthy that our respondents generally favored the three supportive responses over the three negative responses, t (236) = 2.24, p < 0.05 (M = 4.5, SE = .06; M = 3.41, SE = .04 for the supportive and negative responses), and the three negative responses over the two detached responses, t(236) = 6.64, p < 0.001 (M = 2.67, SE = .06 for the detached responses). Our main interest, however, was the interaction between the various FRCNE types and the gender and intensity variables.
Contrary to our hypothesis, we did not find a significant two-way interaction between FRCNE and child's gender (H6), F(7,1652) = .26, p = .96. n.s. ηp2 = 0.00. Thus, fathers did not tend to respond with more supportive FRCNE to girls and with less supportive FRCNE to boys. supportive (M = 3.59, SE = .04; M = 3. 71, SE = .04; M = 3.63, SE = .03, t(236) = 2.41, p < 0.05 for low versus high intensity). A two-way disordinal interaction between FRCNE and intensity was found (H5), F (14,3304) = 26.19, p < .001, ηp2 = 0.11. Fathers' responses to low-intensity questionnaire items tended to be more supportive while responses to high-intensity items were more likely to be negative or detached. Responses to the medium-intensity items were inconsistent. Therefore, for each FRCNE tendency we compared responses only for the high-and low-intensity items.
As presented in Figure 3, respondents were more likely to report using a problem-focused response for the items reflecting low emotional intensity compared with the items reflecting high intensity, t (236) = 6.36, p < 0.001. Likewise, respondents were more likely to report an emotion-focused-comfort strategy for the low-intensity situations, t(236) = 4.47, p < 0.001. However, they were more likely to choose an abreaction strategy (i.e., helping the child express his feelings) for high-intensity items compared with low-intensity items, t(236) = 7.45, p < 0.001. At the same time, the three negative strategies were more likely to be reported for high-intensity relative to low-intensity situations (punitive-aversive, t(236) = 9.11, p < 0.001; punitive-hostile, t(236) = 4.97, p < 0.001; and minimization, t(236) = 3.05, p < 0.005). The same was true for one of the two detached strategies, withdrawal, t (236) = 4.97, p < 0.001. Interestingly, the other detached strategy, giving in, was reported more often for low-intensity items relative to high-intensity items, t(236) = 3.53, p < 0.001.
The results of Study 1 revealed that parental sense of competence mediates the relation between resilience and negative or detached FRCNE. However, contrary to our expectation, no relationship was found between either PSOC or resilience and positive FRCNE. This seems to imply that different variables underpin supportive FRCNE. In Study 2 we took a preliminary step toward examining such potential variables by looking at two attitudinal and behavioral modalities, namely empathy and aggression, and their relationship with supportive FRCNE.

Theoretical background and hypothesis development
The importance of empathy in supportive parenting is well known (Manczak et al., 2016). Empathy is defined as a modality that allows the other person to feel genuinely understood (Magid & Shane, 2017). Davis (2004) treats empathy as a construct encompassing four cognitive and emotional dimensions: perspective taking (an ability to spontaneously adopt the perspectives of other people), fantasy (a tendency to identify with characters in fictional situations), empathic concern (feelings of warmth, compassion, and concern for others), and personal distress (personal feelings of anxious, discomfort and helpless at tense emotional situations, e.g., emergencies). Empathy can negotiate the gap between the self and the other by enabling the self to share the other's felt distress, resulting in increased positive behaviors and decreased hostile or aggressive behaviors (Feddes et al., 2015;Jolliffe & Farrington, 2004).
Much of the research on empathy in parenting has revolved around the relationship between parental empathy and children's emotional and social development, especially among abusive parents. Wiehe (2003) found that abusive parents exhibit lower empathy (perspective taking and empathic concern) relative to foster parents, as well as less self-confidence and less impulse control. These results have been repeatedly supported (e.g., Migliorini et al., 2016;Moed et al., 2017). Wiehe's (2003) findings also support self-efficacy as a foundation for supportive parenting. In keeping, parental empathy has been shown to contribute to the development of children's emotional and social skills, while parental aggression is associated with the development of aggression, emotional difficulties, and social problems in children (e.g., Conger et al., 2003;Yaros et al., 2016).
Based on the foregoing discussion, we can posit that parental resilience and sense of confidence by themselves are insufficient foundations for supportive parenting, because these attributes do not furnish the parent with a deep understanding of the reasons for their child's feelings and behavior. That is, while parents with high resilience and PSOC alone, but low empathy, can direct themselves to think positively about their relationship with the child, it is empathy which provides a window onto the child's emotions, and therefore enables parents-in our case, fathers-to exhibit more positive FRCNE.
As described above, the situations in our study capture cases in which the child's anger and frustration are triggered by the father's own behaviors and decisions. These situations challenge fathers' ability to cope with children's negative emotions when the father himself is the target of those feelings. A sense of being targeted or attacked is known to elicit feelings of aggression (Lobbestael et al. (2016). As such, it follows that fathers who provide supportive FRCNE may need to overcome their inner intuitive aggressive reaction to being the target of negative feelings from their children. If no relationship is found between fathers' aggression and positive FRCNE, this may suggest that fathers who react with positive FRCNE do not perceive the situation as hostile. Based on the foregoing we propose the following -Hypothesis 7: Parental empathy-perspective taking, fantasy and empathic concern -will be positively associated with supportive FRCNE, however, personal distress will be negatively associated with supportive FRCNE. Parental empathy will explain the variance of supportive FRCNE better than parental resilience or PSOC.
Hypothesis 8: Parental aggression will not be associated with supportive FRCNE.

Participants and procedure
Seventy fathers participated in an online survey in exchange for approximately 10. USD To ensure that the samples for the two studies would not overlap, fathers who participated in Study 1 were not sent the link for Study 2. The criteria for participation were the same as in Study 1. Participants in Study 2 were over all similar to those participated in study 1. Their mean age was 46.10 years (SD = 5.5), 69 were married and 1 divorced, and 94% of them lives in cities (pop.>250,000) while the rest in towns (pop.<250,000). Participants earned average wages in Israel. The mean number of children in the family was 2.17 (SD = .43). Participants (beside one) had completed their high school education M = 14.90 years (SD = 2.41). 54% of the participants related to their boys while 46% to their girls, t = 0.33, p = .74, n.s.
Participants completed a survey comprising five scales, including the three used in Study 1-Fathers' reactions to children's negative emotions questionnaire, Resilience scale and Parenting sense of competence scale-and two additional scales measuring empathy and aggression.

Measures
Empathy was measured using Davis's (1983) well-known Interpersonal Reactivity Index (IRI). This is a 28-item self-report questionnaire consisting of four 7-item subscales, each of which assesses a specific aspect of empathy. The Perspective Taking (PT) subscale measures the tendency to adopt others' point of view in everyday life (e.g., "I believe that there are two sides to every question and try to look at them both"). The Fantasy (FS) subscale measures the tendency to transpose oneself into the feelings and actions of fictitious characters in books, movies, and plays (e.g., "I daydream and fantasize, with some regularity, about things that might happen to me"). The Empathic Concern (EC) subscale measures the tendency to experience feelings of warmth, compassion and concern for other people (e.g., "I often have tender, concerned feelings for people less fortunate than me"). The Personal Distress (PD) subscale taps one's own feelings of personal unease and discomfort and helpless at tense emotional situations (e.g., "I tend to lose control during emergencies"). Items are rated on a 5-point Likert scale where 1 = "does not describe me well" and 5 = "describes me very well". Cronbach's alpha was .81 for the full IRI scale and .72, .75, .78 and .82 for the PT, FS, EC, and PD subscales, respectively.
Aggression was measured using the Reactive-Proactive Questionnaire (RPQ) of Raine et al. (2006). This is a 23-item self-report questionnaire divided into two subscales covering reactive aggression (12 items) and proactive aggression (11 items). All the items were rated on a 5-point Likert scale (1 = "never", 5 = "very often"). Internal consistencies in the present study were α = .94 for the reactive aggression subscale, α = .86 for the proactive aggression subscale, and α = .93 for the total RPQ scale. Table 3 presents the intercorrelations between the study variables. All results are compatible with H7 and H8. As expected (H7), empathic concern-one of the four subscales of empathy-was highly positively correlated with supportive FRCNE while aggression (H8) was not. The other empathy subscales were also positively correlated with supportive FRCNE, though to a lesser degree than empathic concern.

Results
To farther explore these hypotheses, we conducted a hierarchical multiple regression exploring the explained variance of supportive FRCNE. The demographic variables were entered in the first step, following by the child's gender in the second step, then resilience and PSOC in the third step, and empathy and aggression in the final step. As expected, when adding empathy subscales variables they were significantly associated with supportive FRCNE and significantly explained the variance of supportive FRCNE. More importantly, among the empathy subscales, empathic concern explained 19% of supportive FRCNE, while the other three subscales (perspective taking, fantasy, and distress) were not found to significantly explain the variance of supportive FRCNE. As expected, aggression did not explain any of the variance of supportive FRCNE (see Table 4).

General discussion
When frustrated it is likely that children are going to react with anger and resentment to some of their parents' childrearing and disciplinary decisions (e.g., with respect to curfews or expensive purchases) or demands (e.g., that the child clean his room or change into a clean shirt). Providing children with a positive and supportive response in the face of protests and aggression is one of the great challenges of parenting, and one that is both cognitively and emotionally demanding, especially when that aggression and frustration is directed toward the parent. From an evolutionary perspective, when one is attacked the intuitive response is to either run away or fight back (Marx et al., 2008). However, from a parental, social and personal perspective, a supportive response is far better for the child's development. Supportive responses, whether problem-focused or emotion-focused, strengthen the child-parent relationship (Cassidy, 1994) and lead to better outcomes for the child's emotion management and regulation (Dumont & Paquette, 2013;Power, 2004); social skills, including the ability to comfort and help others; and school achievement (Eisenberg et al., 1999). Against this backdrop, this study sought (a) to identify parental abilities and attributes that help fathers respond to their children's negative emotions with a positive and supportive parenting; and (b) to examine whether the child's gender and the intensity of the child's emotional display might influence the paternal response in these situations.

Indirect association between resilience and FRCNE through PSOC
Abundant research on parenthood reveals that in recent decades, authoritarian parenting practices have been falling away in favor of more egalitarian family environments. In place of the stereotypical Western family of yesteryear, in which fathers were perceived as decision-makers and mothers as caregivers, both parents now share decision-making and caring roles (Hallers-Haalboom et al., 2015). In addition, childrearing goals and practices have changed. Parents increasingly allow children to express anger and frustration toward them (Trifan et al., 2014).  Meanwhile, new models of parenting stress flexibility in parental control and monitoring, with parents adjusting their vigilance based on the signals emitted by children (Omer et al., 2016).
Against this backdrop, we propose that parental sense of competence is an essential attribute that allows parents in general to provide a supportive parental response in the face of situations such as those discussed in this paper. Likewise, we highlight resilience as an internal quality that helps individuals "bounce back" (Windle et al., 2011) after expending cognitive and emotional energy, and that can help give fathers the sense of competence in their parental role needed to support children through their own challenges.
Our results show, first, that both resilience and PSOC are negatively correlated with unsupportive (i.e., negative and detached) FRCNE. Second, they suggest that the association between resilience and both forms of unsupportive FRCNE is indirect through PSOC. These results support the Bandura self efficacy model (Bandura, 1997), which holds that parents (in this case, fathers) with low power and a low sense of efficacy feel helpless coping with children's aversive behaviors (Bugental, 2009). The current results thus support previous studies which found that as fathers' self-efficacy increases, so does paternal involvement in childrearing (Kwok et al., 2013) and paternal affection toward children (Day & Lamb, 2004), while children's health problems decline (Salonen et al., 2009). However, the current study went one step further in stressing the role of paternal self-efficacy not only in the day-to-day care of young children, as in most previous studies, but with young adolescents whose negative behavior is directed toward the father.
Our results and conclusions may illuminate other findings within this domain. Nelson et al. (2009) found that family stressors such as marital dissatisfaction are negatively correlated with fathers' supportive responses to children's negative behavior, and positively correlated with non-supportive paternal responses. The current findings may also suggest that fathers' self-efficacy is crucial whether stress comes from within the father-child relationship, or from exterior relations that effect the father-child relationship.

Empathy and positive FRCNE
Another important finding of the current study is that whereas resilience, PSOC and aggression were not associated with supportive FRCNE, empathic concern was significantly correlated and, was the only variable able to explain alone nearly 20% of the variance of supportive FRCNE. These findings suggest that resilience and parental sense of competence are necessary parental and personal abilities for self-restraint when a negative or detached paternal reaction may be presented, but empathic concern is required for a supportive responses. Thus, fathers with strong resilience who also have high levels of empathy not only understand abstractly that hitting back or shouting (or alternatively, failing to engage with or giving in to their children) will not bring about lasting change in children's negative behavior; they are also able to moderate their own behavior and reactions accordingly. In addition, as expected, aggression-either proactive or reactive-was not correlated with supportive FRCNE or with any empathic concern. Therefore, it is likely that fathers who react with supportive FRCNE are able to set aside any natural, intuitive inclination to perceive negative behavior directed toward themselves as hostile. These results coincide with other findings regarding conflict resolution behavior among parents and adolescents, which found that empathic concern reduces conflict escalation (Van Lissa et al., 2016).
To sum up the foregoing, the findings suggest that in general, even if parents have a strong emotional connection with their children, coping with particularly challenging situations where children aggressively protest their parents' decisions requires resilience, followed by empathic concern. The latter (i.e., empathy) promotes parents' understanding that as much as their children's protests are inconvenient and unpleasant, such behavior is actually a legitimate reaction when one's wishes are contradicted. By helping the parent see the situation from the child's point of view, empathic concern serves as an essential foundation for a more supportive paternal reaction.

Gender and FRCNE
The current results did not show a greater tendency among fathers to use supportive FRCNE toward girls than toward boys, or vice versa for negative FRCNE. These findings support those of Hurrell et al. (2015). In contrast, Perry et al. (2015) found that, based on adults' childhood memories, mothers tended to react more punitively toward negative emotions in their sons than in their daughters. Cabrera et al. (2012) also found differences in how both parents treated children of different genders. However, their study looked particularly at households marked by depression, conflict, or a chaotic family environment, whereas the current study focused on daily childrearing and disciplinary decisions.
Previous differences found in FRCNE toward sons and daughters may also be explained by looking at other findings about differences between boys and girls. For instance, young girls relative to boys have been found to show less aggressive behavior (Rubin & Burgess, 2002), greater social competence, and higher levels of guilt, empathy, and social-cognitive maturity (Hoglund & Leadbeater, 2004;Kochanska et al., 2002;Olweus & Endresen, 1998). It may be easier for fathers to engage with children who are less aggressive and more competent with more supportive FRCNE and less negative FRCNE. However, the current results may also reflect contemporary family environments, where power tends to be decentralized and is shared relatively equally by both parents regardless of the sex of the children.

Moderating effect of children's emotional intensity on FRCNE
The relationship between the intensity of children's emotional expressions and parental reactions has not been extensively studied. A review of the literature reveals only one study dealing with the relationship between children's emotional intensity, as rated by blind observers during indoor and outdoor free play, and parenting behavior (Fabes et al., 2001). Those authors found a relationship between harsh parental coping strategies and the intensity of children's negative emotions. However, in that study, the source of the child's emotion was exterior to the child-parent relationship. Thus, to our knowledge, the current study is the first to manipulate the intensity of the child's negative emotion in hypothetical situations where the source of the child's frustration is within the family, not exterior to it.
The current results show that low-intensity emotional displays were associated with more supportive paternal responses, while high-intensity displays were met with negative paternal reactions. Reflecting on the self-efficacy model, intense emotions in children may challenge parental coping skills. Helping children to express their feelings (abreaction), while avoiding aversive punishment, requires fathers to exhibit high parental functioning skills in a challenging environment. Our findings recall similar results showing that fathers are less responsive (Corapci & Wachs, 2002) or more likely to employ negative coping strategies (e.g., physical punishments, such as spanking) for boys when the family environment becomes disorganized and chaotic (Cabrera et al., 2012).

Final remarks
The current research was concerned mainly with anger exhibited by children toward their fathers as a reaction to mundane paternal childrearing and disciplinary decisions. Study 1 and 2 have yielded support for the proposed conceptual model suggests that in the face of child frustration caused by paternal child rearing daily decisions. paternal resilience and competence are essential to reduce hostile parenting reactions toward the child, while paternal emotional attributes will increase positive parenting responses. On the basis of the present findings, several promising research directions suggest themselves. First, further research could study fathers' reactions in relation to other feelings experienced by children, such as sadness or fear. Such emotions might stimulate fathers' compassion, and might lead to a different interaction between the parental response and the intensity of the child's emotion. Second, further studies could enlarge our knowledge regarding parental abilities such as empathy and resilience that may help fathers react in a positive and supportive way when children exhibit negative feelings in response to mundane disciplinary decisions. In particular, from a practical point of view, future work should explore ways by which practitioners-or the larger society generally -can help fathers to strengthen and develop their empathy and resilience. Third, unlike some previous findings, we found no difference in how fathers relate to their sons and daughters. This may reflect a change in contemporary family environments that warrants further exploration. Finally, from a slightly different point of view, the current study assumed that fathers are willing to set limits on their children. However, in some cases parents may avoid setting limits or imposing discipline not only to avoid setting off their children's protests, but because it is emotionally complicated for fathers to entertain the thought that their children may harbor enmity toward them. This possibility calls for studies aimed at delineating these two potential emotional paths.
In closing, we must highlight a few points that limit the generalizability of the current findings. First, we focused only on fathers, and so interactions with mothers' parenting skills cannot be deduced from the current results. Second, the sample comprised Israeli fathers, from a culture that has been described as "child-oriented" (Lavee & Katz, 2003). This may impede generalizability to families in other sociocultural contexts (Scharf, 2014). Third, resilience in the current study is presented as a personal attribute. However other scholar's propose resilience is both a process and an outcome being the capacity of a system for successful adaptation to disturbances that threaten system function, viability or development (Masten, 2016). Forth, the current results were based mainly on very modest and very intense children's reactions. Future studies must focus on reactions of moderate intensity before such cases can be represented reliably. Fifth, this study is based on father's self report. As such is subject to social desirability and to the possibility of shared variance. Spontaneous direct observation by further studies on the father-child situations listed above could confirm the findings reported here. Nonetheless, the present findings produce a clear picture of the role of paternal resilience, sense of competence, and empathy in helping fathers cope with children's aggressive protests against paternal childrearing decisions.

Citation information
Cite this article as: Parental sense of competence, resilience, and empathy in relation fathers' responses to children's negative emotions in the context of everyday paternal childrearing decisions, Ido Ziv, Nadia Golbez & Nili Shapira, Cogent Psychology (2020), 7: 1794681. Notes 1. The difference between reaction, a quick, intuitive, sometimes tense and aggressive action as opposed to response a thoughtful, slow and calm action is important to the issue of parenting and as such to the current research that study child rearing practices. 2. Religiosity in Israel is usually divided into three categories -secular, traditional, and orthodox/religious, where "secular" implies little or no religious faith or practice; "traditional" implies minimal religious faith or practice, mainly at home; and "orthodox" implies extensive religious faith or practice, both at home and in public contexts. In reality there is a fourth category, namely ultraorthodox (Haredi in Hebrew). However, the Haredi category is not typically used in surveys because ultraorthodox Jews live extremely insular lives, avoid use of the Internet and other trappings of modern life, and restrict their interactions with people outside their community. To maintain a cautious approach, the current sample included only participants who self-defined as secular or traditional.