Family Socialisation Styles and Hostility in the Adolescent Population

: The affective involvement of parents in the socialization of their children is fundamental for the proper psychological and emotional adjustment of adolescents, although we know that it is difﬁcult to study. In this research, the relationship between parenting style and hostility was analysed in Spanish adolescents. Five-hundred and thirty-six adolescents participated in this study (53.7% males and 46.3% females), between the ages of 12 and 18 (M = 15.76, SD = 1.43), enrolled in 4 compulsory Secondary Education centers of the Castellon province. Family socialization was evaluated through the parental styles: authoritative, indulgent, authoritarian and negligent, and hostility, through the Cook–Medley Hostility Scale Criteria assessment. A multivariate analysis of variance (MANOVA) was carried out with multiple regression analysis. The results showed that hostility and compound hostility are higher in adolescents from authoritarian and authoritative families than in adolescents from indulgent families. It was also observed that the Mother’s Strictness/Imposition had the greatest effect on hostility, as well as a low acceptance/involvement of the father. Regarding parental styles, it was observed that mother’s deprivation, physical coercion, and mother’s verbal coercion were the greatest predictors of hostility. No differences were observed depending on the gender.


Introduction
Current studies emphasize that family environment and parental styles are the main tools available to young people to achieve a correct social adaptation. Regarding the social climate of the family, effective communication allows an appropriate adjustment of adolescents [1,2]. From the Pinquart study [3], we know that the authoritarian, permissive, and neglectful parenting styles are associated with higher levels of externalizing problems, and that victimization and deviant behavior are influenced by negative control [4].
Parenting style can be defined as " . . . a constellation of attitudes towards the child that, considered jointly, create an emotional climate in which the parents' behaviours are expressed" [5] (p. 116). Parental support and control are the two factors that explain most of the variability in the disciplinary behaviour of parents. Two dimensions have traditionally been considered in parental socialization: responsiveness and demand, also called acceptance/implication and severity/imposition [6][7][8] to explain the parental styles. This classic two-dimensional model is one that best explains the parental actions of socialization, despite the different names that different authors use. Thus, acceptance terms [9], attachment [10], love [11][12][13] have been used for the acceptance/implication axis, and for severity/imposition, those of domination, hostility, rigidity, control, or restriction are used [10][11][12][13]. The acceptance/implication dimension refers to the degree to which parents show affection to their (anger), which is influenced by a negative attitude towards others (hostility) and results in violent behaviour or aggression.
Nevertheless, Barefoot [41] had already suggested that, given that hostility often (but not always) involves affective, behavioural, and cognitive aspects, the three of them should be articulated and evaluated separately.
The affective component of hostility includes a variety of emotions such as anger, resentment, and annoyance. The cognitive component is made up of negative beliefs about human nature that include cynicism (widespread mistrust of others), hostile attributions (malice, antagonism, and threat), negative expectations (mistrust or suspicion), and a negative attitude (enmity, negativism and resentment). Lastly, the behavioural component includes aggressiveness, antagonism, negativism, and insults.
All these characteristics suggest that hostile individuals have greater difficulty in their interpersonal relationships, a higher number of conflicts and less social support, which, generally speaking, results in higher psychosocial vulnerability. In this regard, the absence of social support and excessive hostility seem to increase the risk of heart disease and deaths related to heart disorders. Other studies show that high scores of cynicism maintain and increase cardiovascular and neuroendocrine reactivity towards stressors. This situation also implies a greater predisposition to elicit and exacerbate interpersonal conflict in their lives, which reduces social support and thus maintains their hostile beliefs and behaviour towards others. Lastly, their permanent state of vigilance with respect to their surroundings means that the physiological changes associated to that state can become chronic [42].
It is within this framework of research on hostility and parenting styles that we propose our general aim of analysing the relationships between parental socialisation styles and hostility in teenagers. Previous studies conducted on the family have taken into account the educational dimensions of expression of affection and control, and their relationship with children's adjustment. The indulgent parenting style is the one that is most closely related to psychosocial adjustment in adolescents [25]. Two hypotheses are proposed: (1) the indulgent parenting style is related to lower scores on hostility, and (2) the authoritarian parenting style is related to higher scores on hostility. In numerous studies it has been found that the most negative style, due to its effects on children, is the negligent one, followed by the authoritarian one. The most functional have been, in this order, the indulgent and the authoritative [32,37,[43][44][45]. This research has not formulated this hypothesis related to authoritative or negligent styles because it was not among the objectives of this research to analyse these effects.

Participants and Procedure
The sample in this study consisted of 536 participants, 53.7% males and 46.3% females (Mean = 15.14 years old, and SD = of 1.95). The age range was 12-18. In Table 1, the number of subjects by age is shown and in Table 2, the number of subjects per age group.
The study was conducted in four public secondary schools in the province of Castellón, Spain. To this end, permission was obtained from the Headteacher and the Regional Secretary of Education. This research was authorized by the ethics committee of the University of Castellón, dated 30 March 2015. All students took part in the study on a voluntary basis and all parents signed informed consent. The questionnaires were answered individually and anonymously in a series of tutorial sessions each lasting around 50 min, under the supervision of two specially trained professionals. The schedule for the groups was designed according to terms throughout the academic year 2015-2016, depending on the timetable of each tutorial, by levels and educational stages, that is, ESO (compulsory secondary education), high school, and middle-and upper-level Vocational Training.

Instruments
Cook-Medley Hostility Scale [46]. Ho was validated for the Spanish population [47], and Cronbach's alpha value for this scale was 0.80, This scale consists of 50 items taken from the Minnesota Multiphasic Personality Inventory (MMPI), which must be answered with a True-False response, grouped in six subscales: cynicism, hostile feelings, aggressive answers, hostile attribution, social avoidance, and others. The following are examples of the questions that measure hostile responses: I have often been given orders by someone who didn't know as much as me. People often disappoint me. I think I've often been punished without reason, etc. The cynicism, hostile feelings, and aggressive answers scales make up the Composite Hostility (CHOST) measure or the more cognitive dimension of the inventory [48], which alone or in combination with defensiveness (that is to say, the tendency to suppress socially unacceptable aspects of oneself), is considered as the most important criterion for predicting cardiovascular reactivity.
The internal consistency for the hostility questionnaire was calculated by means of Cronbach's α, which yielded a value of 0.791 for the full scale; 0.730 for the Compound Hostility (CHOST) scale (Subscales: cynicism, hostile feelings, aggressive answers); and 0.696 for the cynicism scale.
Parental Socialisation Scale (ESPA29) [49]. This scale is based on the bi-dimensional theoretical model of parental socialization [7]. ESPA29 was validated for the Spanish population [50], and Cronbach's alpha value for this scale was 0.95 and 0.93 for this study. The reliabilities of the dimensions ranged between 0.90 and 0.93 for both parents. It evaluates the parents' socialisation styles in different natural scenarios that represent the daily life of families in western culture. It consists of 232 items, 116 devoted to the father figure's way of parenting and another 116 referring to that of the mother figure. The child rates the parenting of his or her father and mother separately in 29 situations (with four possible answers: 1-never to 4-always), 13 of these situations referring to the obeying of family rules and 16 concerned with disobedience. A global measure is obtained in two dimensions: acceptance/involvement and coercion/imposition, which are similar to the responsiveness and demandingness previously proposed [7]. The family score on the acceptance/involvement dimension was obtained by averaging the answers on affection (He/she is affectionate with me), dialogue (He/she talks to me), indifference (He/she appears indifferent), and apathetic (He/she doesn't care). The scores on the last two subscales were obtained by inverting the answers, as they were negatively related to the dimension. The family score on the coercion/imposition dimension was obtained by averaging the answers on deprival (He/she deprives me of something), verbal coercion (He/she scolds me), and physical coercion (He/she hits me). The scores on the two dimensions were used to classify the parenting style of each parent as authoritative, indulgent, authoritarian, or neglectful, the sample being divided by the median on each axis of the bi-dimensional model. An average reliability of 0.90 was obtained for the scales of the instrument, namely: overall scale, scale for the father, and scale for the mother, as well as for their dimensions: acceptance/involvement and coercion/imposition. Acceptance/involvement dimension: CR = 0.971. In both instruments, a back translation was used. Our scores are similar to the normative samples of both measures.

Data Analysis
First of all, all scales' reliability was analysed from Cronbach's α. Secondly, the Parental Socialisation Questionnaire (ESPA29) was categorised by classifying the families according to their parenting style (simultaneously bearing in mind two dimensions of the socialisation model, Acceptance/Involvement and Strictness/Imposition) by dichotomising the sample by the median (percentile 50) in the two dimensions. Thus, authoritative parental style scored above the median on both dimensions; indulgent parental style scored above the median on Acceptance/Involvement and below it on Strictness/Imposition; and authoritarian parental style scored below the median on Acceptance/Involvement and above it on Strictness/Imposition. The Neglectful parental style scored below the median on both dimensions. Thirdly, a multivariate factorial analysis (MANOVAs) were applied with the criterion variable hostility, the parenting style (authoritative, indulgent, authoritarian, and neglectful) and sex of adolescents (male and female) being considered as independent variables to test for possible interaction effects. Univariate F tests were later carried out to examine the differences in the variables analysed and the Bonferroni post-hoc test was applied. Finally, a multiple regression analysis was carried out.

Results
A first approach to the data was to calculate the number of fathers and mothers integrated in each style of socialization based on the perceptions of the children. In Table 3, the distribution of mothers according to socialization styles is shown. Although the distribution is relatively homogeneous in all styles, it is striking that the predominant styles are negligent and authoritative, in this order. The same analysis was carried out with respect to the father and a distribution similar to that of the mother was observed with a minimal differences (Table 4).
MANOVA (4 × 2) design, considering the Parenting Styles (A), (a 1 , indulgent; a 2 , authoritative; a 3 , authoritarian; and a 4 , neglectful), and the Sex of the adolescent (B), (b 1 , male; b 2 , female) as the independent variable, and hostility as the dependent variable, the results showed statistically significant differences in the main effect of the parenting style, Λ = 0.958, F(9, 1280.296) = 2.540, p < 0.01. The main effect of sex did not show any statistically significant differences, Λ = 0.998, F(3, 526.000) = 0.390, p > 0.05, and neither did the interaction effect between parenting style and sex, Λ = 0.976, F(9, 1280.296) = 1.405, p > 0.05. Table 5 shows data.  Moreover, ANOVA was performed to analyse the differences in each of the dimensions of hostility (Hostility, Composite Hostility and Cynicism) taking into account the four parenting styles described above. The ANOVA showed statistically significant differences in the dimensions of Hostility, F(3,532) = 4.464, p < 0.01, and Composite Hostility, F(3,532) = 5.140, p < 0.01 (See Table 6). An analysis was then performed to determine which parenting styles had significant differences between them by applying the Bonferroni post-hoc test to compare the means. The results indicated that the children from authoritarian families obtained higher scores on the hostility dimension than  A separate regression analysis examining the relative contributions of the parental dimensions (as continued measures rather than dichotomized, separately for mothers and fathers) on hostility has been carried out. The Strictness/Imposition and Acceptance/Involvement from both parents was introduced in the analysis as predictive variables and hostility as the criterion variable. The multiple linear regression analysis, step by step, whose results are presented in Table 7, showed that, in the predictive variables of hostility, two were statistically significant: mother Strictness/Imposition (β = 0.238) and father acceptance/involvement (β = −0.095). The standardized regression coefficients β indicate that all the variables have a certain weight on the criterion variable. According to this statement, the percentages of variance explained (adjusted coefficients of determination) for each of these predictor variables were of average magnitude for the two variables, respectively: 4.6% and 5.5%. High Strictness/Imposition of the mother and low acceptance/involvement of the father were predictive variables of hostility, with its explanatory power being medium, since they explain 5.5% of the variance. The interpretation of the β of each predictor variable on the criterion variable, hostility, for example, would be the following: for each unit added to the variable mother Strictness/Imposition, 0.238 points the hostility. The multiple linear regression analysis, step by step, whose results are presented in Table 8, showed that in the predictive variables of hostility, four were statistically significant: mother deprivation (β = 0.161), father dialogue (β = −0.187), mother physical coercion (β = 0.132), and father verbal coercion (β = 0.107). The standardized regression coefficients β indicate that all the variables have a certain weight on the criterion variable. According to this statement, the explained variance percentages (adjusted coefficients of determination) for each one of these predictor variables were of average magnitude for the four variables respectively: 3.6%, 6.3%, 8.1%, and 8.8%. High mother deprivation, low father dialogue, high mother physical coercion, and high verbal coercion father were predictive variables of hostility, with their explanatory power being medium, since they explain 8.8% of the variance. The interpretation of the β of each predictor variable on the criterion variable, hostility, for example, would be the following: for each unit added to the variable mother deprivation, 0.161 points go into hostility.
It was also observed that the Mother's Strictness/Imposition had the greatest effect on hostility, as well as a low acceptance/involvement of the father. Regarding parental styles, it was observed that mother's deprivation, physical coercion, and mother's verbal coercion were the greatest predictors of hostility. No differences were observed depending on the gender.

Discussion
The aim of this research was to analyse the relationships between parenting styles and hostility in adolescents. It has been observed that adolescents from indulgent homes (affectionate and not very imposing parents) are more secure, trusting, open to dialogue, and tolerant of their peers (less hostile), which confirms the first hypothesis of this study. As has been highlighted earlier, very little research has been conducted in the field of hostility and parenting styles. The results are also consistent with previous studies which have highlighted the relationships between the different parenting styles and adjustment in teenagers [17,51]. Additionally, in previous studies, it has been observed that the indulgent style is also predominant in other non-English-speaking countries, such as Brazil [52], Italy [53], Portugal [54], which is, in our opinion, very interesting insofar as it incorporates a relatively new dimension into these results: hostility.
In addition, regression analysis provides some nuances that we find very interesting, such as the fact that the Mother's Strictness/Imposition had the greatest effect on the hostility of the children as well as low acceptance/involvement of the father, which is a novelty in this field of study. Similar results had already been obtained in the area of peer violence in the adolescent period [2,17,25,55], but not in the hostility field. Regarding parental styles, it was observed that mother's deprivation, physical coercion, and mother's verbal coercion were the greatest predictors of hostility. These data enrich the previous results that refer to the Mother's Strictness/Imposition. Regarding gender, although no differences were found in the MANOVA, the results obtained from the regression allow us to start, or move forward, in a very suggestive line of work that can offer excellent results. This would be related to the role that the mother, and not the father, at least not so much, have in hostility and violence. New studies are necessary and if these results are confirmed, it would entail a great enrichment of the prevention and intervention programs, because what they point to is the role of the mother in the prevention of hostility and violence.
Secondly, the hostility and cynicism dimensions have been seen to be the highest in adolescents brought up in families in which the authoritarian style predominates, which therefore confirms part of the second hypothesis in that anger and aggression are higher in the authoritarian style. The third dimension, composite hostility, is however similar in teenagers brought up both in families in which the authoritative style predominates and in those where the authoritarian style prevails. Therefore, children from authoritarian families and from authoritative families, respectively, are the ones who were seen to be more mistrustful and hostile than adolescents from indulgent families. An upbringing that is based on patterns of an imposing nature and the absence of dialogue or the chance to reason fosters hostile attitudes in children, which are related, either implicitly or explicitly, with aggressive behaviour. When children perceive irritability, greater control, and an attitude of rejection in their parents, their aggressive behaviours and their frustration increase because they feel their freedom is being restricted. In the case of authoritative families, expressing affection reduces the unease associated with a high level of control imposed by the parents, although it results in high levels of composite hostility in their children. An authoritarian upbringing is a risk factor for hostile behaviours in youngsters due to the possibility of them reproducing the patterns of irritability, intolerance, and lack of empathy existing within the family context in other contexts, such as at school and/or with a partner [26,56,57].
These results are also consistent with those previously obtained [58] on the importance of family support and protection to foster safe and trusting relationships in children instead of rejection and hostility as promoters of insecure interaction based on expectations of mistrust, aggression, and hostility in their relationships with others. They are also in line with the other results [59] on the direct relationship between hostile behaviour and the use of punitive strategies in controlling the child's behaviour, and other study [60] about negative communication and mistrust. We believe these results are interesting and deserve to be explored in greater depth due to their theoretical and practical implications.
Our results, however, do not support the findings of other studies carried out in English-speaking contexts, in which it is concluded that the authoritative parenting style (high affection and high imposition) is related to better psychological and social adjustment of one's children [17,61]. In this regard, according to our results, Spanish adolescents from authoritative families are the ones that obtain the highest scores on composite hostility, and therefore the binomial high control-high affection does not appear to work properly in the Spanish context, contrary to what occurs in the English-speaking context. In this respect, it is observed that the negative effects of control are not mitigated by affection and responsiveness, so that strictness and excessive control, although accompanied by affection, appear to encourage hostile and malicious attitudes, which could undermine their socio-moral and affective development and give rise to forms of relationships based on mistrust and a lack of empathy from an early age. This is a result that should be explored further from an intercultural perspective and taking into account, more specifically, the variable hostility, because in previous studies mentioned earlier hostility is a collateral dimension.

Limitations
One of the limitations of this study could be the fact that families are classified according to their parenting style, based exclusively on the answers given by their children. Nevertheless, there is empirical evidence showing that children's answers tend to be less biased due to social desirability than the answers given by parents [62]. Finally, the cross-cutting nature of the study does not allow conclusions to be drawn about the causal relationships between the study variables and therefore, as noted earlier, we suggest that longitudinal study designs should be used to test the relationship with, first, cardiovascular pathology, and second, the transmission of violence from one context to another. The ultimate purpose would be to design health-promoting preventive strategies.
Despite these limitations, the results of this study highlight the importance of parenting in adolescents' psychological and emotional adjustment, while also providing empirical evidence for the incipient research being carried out in our country in the field of parenting. The family matters; it is an asset for the adolescent's positive development if it provides support and affection, resulting in a reduction in the frequency of emotional and behavioural disorders [63,64]. In this regard, there is clearly a need to consider family assets in our cultural setting when designing intervention programmes [65] that are aimed at improving educational practices, the processes of parental communication, familial acceptance, and involvement. They should also include, in addition to positive parenting, the development of school and community resources, with the aim of providing the support and experiences needed to foster the full development and adjustment of our teenagers.