How are parental practices and attitudes towards corporal punishment related to child academic, developmental, or psychological–emotional dysfunctioning?

Corporal punishment (CP) is a widely spread disciplining practice among parents and caregivers globally. Our paper aimed to explore the relationship between the parental attitudes towards CP, expected outcomes of CP, and parenting practices on one hand, with the reported dysfunctions of their children, on the other. Additionally, we aimed to explore the relationship between the use of CP and the reported academic, developmental, and psychological–emotional dysfunctions of their children. The present study involved a nationally representative sample of 1186 parents in Serbia, who had at least one child aged 0–18 years at the moment of interviewing. The parents filled out a series of questionnaires on their attitudes towards CP, expectations of CP outcomes, and their parental practices. Findings indicate that parents that report having a child with dysfunctions have positive attitudes towards CP and expect positive outcomes of CP. These parents also report using more CP as a disciplining method, as well as other harsh disciplining practices. We also identified parental positive expectations of CP, use of physical assault, psychological aggression, neglect as significant predictors of reported child dysfunctions severity. Having all the results in mind, we can assume that children with health-related and school-related issues might be at potential risk of further maltreatment.


Introduction
Child discipline is a significant part of child-rearing and the practices used for the purpose of discipline have broad implications for a child's well-being.According to the World Health Organization, it includes training to develop judgment, behavioral boundaries, self-control, self-sufficiency, and positive social conduct [1].However, child discipline can be frequently confused with violent practices such as physical and emotional punishment.Corporal punishment (CP), commonly referred to as "disciplinary spanking", is usually defined as "the use of physical force with the intention of causing a child to experience pain, but not injury, for the purpose of correcting or controlling the child's behavior" [2].It includes behaviors such as hitting, shaking, or throwing objects at a child.Historically, the use of CP was universally present in every culture, traditionally accepted and shaped over time by different personal, religious, social, and cultural values.There was a sharp change in paradigm in the last couple of decades when CP was labeled as a form of abuse against children [3].Despite preventive efforts worldwide [4], epidemiological studies surprisingly show that many parents (up to 80% of parents worldwide) still use 1 3 CP to discipline their children [5,6].This made the use of CP in child-rearing the most controversial topic in the parent-child relations and led to a strong research interest and hyperproduction of studies [7].
The existing body of research on CP shows no evidence of a long-term improvement in child behavior or any aspect of development when physical force is used [3,7].Furthermore, research shows that CP could potentially lead to negative short-and long-term effects on child functioning and mental health.More specifically, the studies report negative effects of CP on behavioral and emotional outcomes [8], internalizing and externalizing behavior problems or cognitive performance [9][10][11][12][13].The harsher disciplinary methods (e.g., severe physical assault) are more strongly associated with the negative outcomes on child's health and functioning [9,14].
Violent psychological discipline practices involve the use of humiliation, insults, guilt or emotional manipulation to discipline the child.The use of this type of discipline also has harmful effects on child's well-being.For example, a study conducted in China reported that children of parents who used violent psychological discipline practices were expressing more aggressive behavior towards others [15].
The existing literature indicates that children with disabilities are more often subjected to violent discipline in comparison to children without disabilities [16,17].This has been a case when parenting a child that has a problem with attention [18], antisocial behavior [19], and physical disabilities [20].The findings also indicate that children with certain disabilities are more subjected to violent practices than others.For example, children with problems in communication (hearing problems or speech problems) are more often subjected to CP than others due to the inability of proper parent-child communication [17,21].However, most of these studies focus on a child's dysfunction in a single domain and do not include dysfunctions in other domains that can be common in children, such as academic functioning and psychological-emotional functioning.
Previous research shows that parenting children with dysfunctions in domains such as mental health can be a significant source of burden and distress for the parent and the whole family [22,23].Consequently, parental distress can be a significant risk factor for the use of violent discipline practices, especially when combined with other factors such as their previous experience of psychological or physical punishment, favorable attitudes towards physical punishment, and lack of sufficient support [24].
Regardless of the unequivocal evidence that indicates that CP can be harmful to a child, many parents still believe in the "benefits" of CP and use it as part of their disciplining practices [6,25].The parental attitudes and norms towards CP can range greatly, from a belief that CP is acceptable and good for the child to the belief that any type of physical punishment is a form of abuse [25,26].The parental attitudes and norms can be associated with a number of factors such as their social surroundings, country of origin, cultural background, and individual factors related to parents and families (e.g., education, social support, socioeconomic status, health, etc.) [27][28][29][30].Back in 1991, Korbin was one of the first authors who highlighted the fact that cultural norms and parental attitudes shape parental opinions on what is child abuse and what is not [31].The research that followed this theory indicated that parental attitudes towards the use of CP can shape their parental practices and the frequency of CP [30,32].In addition, it has been shown that parental expectations of CP effects are an important factor that influences the use of CP.In other words, the parents who expect positive outcomes out of CP, tend to use CP more frequently [26,32].To our knowledge, there are no data in the current literature on possible mediating roles of parental attitudes and expected outcomes of CP when parenting a child with a certain dysfunction.
The primary aim of this research was to explore the relationship between the parental attitudes towards CP, expected outcomes of CP, and parenting practices on one hand, with the reported dysfunctions of their children.In addition, we aimed to explore the relationship between the use of CP and the reported academic, developmental, and psychological-emotional dysfunctions of their children.

Participants and procedure
Having in mind relatively low effect sizes in previous papers, we aimed to determine effects of r = 0.082 or higher, with alpha level of 0.05 and power of 0.8.To do so, the sample included 1186 parents in Serbia, who had at least one child aged 0-18 years at the moment of interviewing.To have representative sample of parents, we used probabilistic multistage sampling, designed as a combination of stratified, cluster and simple random sampling procedure.In the first stage the Republic of Serbia was divided into four main regions, and in the next stage the clusters of 70 municipalities were selected randomly from those regions.These clusters were proportional to the region-size and settlement type (urban/ rural).The final stage was the selection of participants via the random-walk sampling method in each of the selected areas.After entering the household, interviewers informed the household members about the study aims and asked a person whose birthdate is the closest to the interview date to participate in the study.Interviewer would then give an informed consent form to a selected person and ask him/her to sign it prior to interview continuation.From each household, only one member (parent) filled in the questionnaires.
Response rate was higher in rural areas (33.3%) than in urban areas (28.6%).
The data collection process took place in the participants' homes, face-to-face.The questionnaires were delivered by previously trained interviewers through either computerassisted personal interviewing (CAPI) or paper-pencil method.The paper-pencil method was for people who were not comfortable with using a computer for any reason.Participants filled out the questionnaires by themselves and were free to ask questions or any kind of additional help from the interviewers.Interviewers were recently graduated psychologists or students of final year of psychology.
Within the same study, we also collected a sample of 234 professionals, involved with parenting issues though health, social care and education.These participants were collected through probabilistic cluster sampling, with institutions used as clusters.However, data from these participants will not be analyzed in this paper.

Ethics
This study was performed in line with the principles of the Declaration of Helsinki.Approval was granted by the Ethics Committee of the Institute of Mental Health (No1060/2089/1).The informed consent to participate in the study was given to the participants in paper form and obtained prior to taking part in the research.The informed consent forms were kept separately from the questionnaires.Also, the questionnaires were filled out anonymously, thus the confidentiality and privacy of all participants were guaranteed.

General information questionnaire
The general information questionnaire was designed specifically for this study to cover the most relevant sociodemographic information.The first part of the questionnaire assessed the information on the parents (e.g., age, gender, education, employment).The second part assessed parental reports on different dysfunctions that their children are facing.We categorized the problems of children into three groups: (1) Developmental dysfunctions (vision problems, hearing problems, speech problems, intellectual disability, motoric problems, neurodevelopmental difficulties like autism, attention/hyperactivity problems, and learning difficulties).Sample item: "Do you have a child who has speech problems?";(2) Psychological-emotional dysfunctions (aggressive behavior, suicide attempts, nonsuicidal self-injury).Sample item: "Do you have a child who ever attempted suicide?"; (3) Academic dysfunctions (i.e., problems related to school attendance and achievements; unsatisfactory achievement, adaptation problems, truanting).Sample item: "Do you have a child who had adaptation problems in school?".

Conflict Tactics Scales: Parent-Child version (CTSPC)
The Parent-Child Conflict Tactics Scales (CTSPC) is a modified version of the Conflict Tactics Scale (CTS) developed by Straus et al. [33,34].The CTS scale was primarily designed to measure the presence of and degree of maltreatment among partners in a marital, cohabiting, or dating relationship.The CTSPC scale was modified in a way to assess parent-child relationships.The final version of the CTSPC questionnaire consists of 35 items that assess the presence and severity of various forms of parental practices.It is a parent self-report questionnaire that consists of six scales: Nonviolent Discipline; Psychological Aggression; Physical Assault; Supplemental Questions on Discipline in the Previous Week; Neglect; and Sexual Abuse.Each scale contains questions that are scored by the eight-point Likert-type scale that ranges from "once in the past year" to "more than 20 times within the past year".Two responses reflecting the absence of practices in the past year are "not in the past year but it happened before" and "this never happened".All responses that reflected the frequency of practices in the last year were further marked as "last year", while the response "not in the past year but it happened before" was marked as "lifetime".In addition to the overall physical punishment scale, there are three subscales associated with physical assault-the minor physical punishment scale (i.e., CP), the severe physical assault scale, and the very severe (i.e., extreme) physical assault scale.The CTSPC was validated in 2018 by Cotter et al. [35] and the scale showed good internal consistency (Cronbach's α = 0.88).

Attitudes toward physical punishment (ATPP)
The ATPP is a self-report questionnaire consisting of 12 items measuring the parental attitudes and opinions regarding the use of physical punishment as a part of child-rearing.This version of the questionnaire was based on the Discipline Questionnaire (DQ) from the research by Graziano et al. [36].The questions were answered on a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree).Previous studies show good internal consistency (Cronbach's α=0.84) and high test-retest reliability (r = 0.76) [36].

Expected outcomes of using CP
The expected outcomes of using CP are based on the Outcomes of Physical Punishment Scale developed by Durrant et al. in 2003 and modified by Taylor et al. [26].The final scale consists of 14 items, equally distributed in two subscales.The first subscale measured "positive outcomes", while the second scale measured "negative outcomes" of corporal punishment.The positive outcomes included "positive" effects on a child such as correction in short-and long-term behaviors, better family relations, and learning self-control.In contrast, negative outcomes refer to the negative effects on a child's well-being and functioning, such as injuries, abuse, aggressive behavior, and physical and mental health consequences.Participants rated each item on a 5-point Likert scale ranging from 1 = never to 5 = always.For each subscale, an average score was calculated for each set of items.According to previous research, the two subscales showed good internal consistency; Cronbach's α for expected positive and negative outcomes were 0.88 and 0.89, respectively.

Data analysis
The data analysis included descriptive methods (frequencies, percentages, means, and standard deviations), testing differences between groups (Student's t-test), and multivariate analysis (linear regression).All analyses were conducted in SPSS software, version 20.

Results
The socio-demographic data of the respondents in the sample are presented in Table 1.The majority of the sample were mothers (N = 755; 63.8%).About half of the respondents finished secondary school, whereas 44.3% had higher levels of education.The mean age of the participants was 38.08 (SD = 8.06).
The descriptives of the reported dysfunctions of children are presented in Table 2. To analyze the parental attitudes, norms, and practices in accordance with the reported child dysfunctions, we first split the sample into two major groups; the first group was composed of parents who reported having a child with dysfunctions in any area (e.g., developmental, psychological-emotional, or academic) while the second group was composed of parents who reported having a child without any dysfunctions.To obtain more specific results, the first group was further divided into three subgroups according to the reported problems the children were facing: (1) Developmental dysfunctions; (2) Psychological-emotional dysfunctions; (3) Academic dysfunctions (i.e., dysfunctions related to school attendance and school achievements).Each group of parents having a child with certain dysfunctions was compared separately with the group reporting no dysfunctions, instead of comparing all three dysfunction types with the no-dysfunction group at once, since different dysfunction types appeared conjointly and therefore those three groups overlapped.The developmental dysfunctions were reported for every fifth child in our sample.The most common developmental dysfunctions reported were vision problems, followed by speech problems (Table 2).Out of all parents, 18% (N = 213) reported having a child with one developmental dysfunction, 1.8% (N = 21) reported two developmental dysfunctions, while 3 parents (0.3%) reported having a child with three developmental dysfunctions at the same time.
The psychological-emotional dysfunctions were reported in almost 4% of our sample.The parents mostly reported having a child who was aggressive or was engaged in nonsuicidal self-injury (NSSI).Having a child with only one psychological-emotional dysfunction was reported by 3% (N = 41) of the parents, while conjoint psychological-emotional dysfunctions in children were reported by 0.3% of the parents (N = 3).
Academic dysfunctions were most frequently reported.The most common reported academic dysfunction was related to problems in adaptation to school or other educational institution (crying in class, willingness to be with family members, etc.).Having a child with one school dysfunction was detected in 18.7% of the sample (N = 222), two academic dysfunctions in 5.8% (N = 69), and three conjoint academic dysfunctions were reported in 2.1% (N = 25) children.

Expected outcomes of CP
The two groups of parents with and without the reported dysfunctions in children were compared in how much they expected positive and negative outcomes of CP.The t-test for independent samples showed a statistically significant difference between the two groups (Table 3).The parents who had children with dysfunctions more often reported that they expected that the outcomes of CP were positive (e.g., improvement in child behavior, increase in respect towards a parent, learning correct behavior, etc.).We, then, compared the expectations of CP in groups of parents who reported having a child with each dysfunction and those who did not report it, for each of the three types of dysfunctions.The groups of parents with each type of child dysfunction reported believing more in positive outcomes of CP (Table 3), whereas no differences were found when it comes to expectations of negative outcomes.

Attitudes toward CP among parents
We examined the differences in attitudes towards CP between the groups of parents of children with and without any reported dysfunctions.The differences in attitudes towards CP are shown in Table 4.The parents who report having a child with dysfunctions have significantly more positive attitudes toward CP than the parents of children who did not report dysfunctions.Additionally, each group of parents was examined according to the dysfunctions of their children (Table 4).When compared with parents of children without reported dysfunctions, there was a statistically significant difference in attitudes in parents who reported their children having academic and psychological-emotional dysfunctions.There were no differences in the attitudes among parents with children with and without developmental dysfunctions.

Parental disciplinary practices
The results regarding the differences in parental practices between the parents of children with and without reported dysfunctions are presented in Table 5. Parents who reported having a child with any dysfunction engaged more in various types of practices compared to parents without such reports (Table 5).Parents who reported having a child with dysfunction were significantly more engaged in non-violent practices, as well as in psychological aggression, both in the past year and during their lifetime.When it comes to physical aggression, these parents were also engaged in physical assault in general during their lifetime, and more precisely, in minor and severe physical assault, whereas there were no differences in the use of extreme physical assault during their lifetime.However, physical assault, in general, was not differently used in the past year by either group of parents.Thus, further comparison of three different types of physical aggression (minor, severe, and extreme) in the past year was not performed.

Parental practices and positive expectations of CP as predictors of reported child dysfunctions
The next step was to investigate whether the parental disciplinary practices, parental attitudes towards CP, and expectations of positive CP outcomes can be used to predict the severity of reported child dysfunctions for all previously mentioned dysfunction types.The multiple linear regression analysis revealed the significant predictors for each type of reported dysfunction in children (Fig. 1).The positive bars represent the positive regression coefficients (β coefficient), meaning that a higher bar value for each predictor is associated with a higher number of the reported child dysfunctions in the analyzed category.In Fig. 1 we presented only significant predictors for each category.The first regression model showed that 4.2% of the severity of developmental dysfunctions variance in children can be statistically significantly predicted (F(19;1156) = 2,676; p < 0.001).The significant predictors for this type of dysfunction were the expected positive outcomes of CP (β = 0.111; t(1156) = 2.78; p < 0.05) and severe physical assault (β = 0.098; p < 0.05).

Discussion
The most common reported child dysfunction among the participants was the academic dysfunction, reported by every fourth parent (27.2%).The other studies that researched the behavioral dysfunctions in school-aged children report that the prevalence of behavioral difficulties in school-aged children ranges roughly between 23.5% in preschool children [37], and 36.4% in children in elementary school [38].
The second highest reported problem in children was developmental dysfunction which was reported by every fifth parent (20%).The reported data from the studies conducted in other countries show that the prevalence of developmental dysfunctions based on a parental report, ranges from 11.36% in Taiwan [39] to 18% in the United States [21].A review done in 2007 in low-and middle-income countries reports child disability prevalence from 0.4 to 12.7% depending on the study and assessment tool [40].However, the authors state that the rates can be higher due to the problems in identifying the disabilities and the lack of adequate instruments.The most prevalent developmental dysfunctions in the present study were the problems with vision, followed by speech problems and motoric problems.Similar results were found in a recent study done by Chen et al. [39].
The percentage of parents that reported that their child has psychological-emotional dysfunctions was the lowest (3.9%).In Serbia, the psychological-emotional dysfunctions among children and adolescents are largely unrecognized and stigmatized.This is especially the case in rural areas [41].This lack of knowledge with high rates of stigma is probably due to the lack of experts, services, education, and systemic support in the field of child and adolescent psychiatry [41].This can also be the possible explanation for a high percentage of reported academic dysfunctions, and in contrast, a low percentage of reported psychological-emotional dysfunctions.The occurrence of NSSI reported by parents in our study was 1.5% of parents, while suicidal attempts were reported in 0.4%.This finding is in concordance with the Bella study performed by Resch et al. that reported the prevalence rates of 1.4% for self-mutilation and/or suicidal attempts, also reported by parents [42].The data in the literature are mainly based on the child reports, and those studies report higher rates of suicidal behavior [43,44], and higher rates of suicide attempts in youth with a lifetime prevalence of 6% [45].
In the present study, the findings indicate that there are significant differences between the parents that report having a child with and without dysfunctions in terms of their attitudes towards CP, expected outcomes of CP and in using different disciplining methods.
Firstly, the parents of children with any reported dysfunction, as well as with any dysfunction separately (developmental, psychological-emotional or academic), more often believe that the outcomes of CP on children are positive in comparison with the parents of children without reported dysfunctions.In other words, former parents believe that CP will benefit their child's personality and behavior and lead to a behavioral correction, better family relations, and learning self-control [26].One of the first studies that assessed the perceived parental outcomes of CP was done in 2003 by Durrant et al. [30].This study reported that the two samples of mothers in two countries (Sweden and Canada) expected that the outcomes of CP are mainly negative (increased child aggression, longterm emotional upset and parental guilt).A similar finding was also shown among professionals that work with children, where the professionals also believed that the outcomes of CP were primarily negative [26].This is in line with the findings performed in our country where we analyzed the differences between the expected outcomes of mothers and fathers [25].More specifically, mothers or female caregivers were more often reporting the use of violent psychological and physical practices compared to fathers or male caregivers.However, mothers and female caregivers were also more frequently engaged in developmentally supportive activities.This is in line with other studies that report that mothers use CP and other violent practices more often than fathers do [46][47][48].
The studies that measured perceptions of parents about CP did not include the perceptions of parents that care for the child with dysfunctions.To our knowledge, there is no previous literature on expected outcomes of parents of a children with dysfunctions.From the qualitative study done by Whittingham et al. [49] the responses of parents outlined some possible factors that shape their opinions on this matter.Some of these factors included difficulties to determine which behaviors are out of the child's control versus which are amenable to change [49].Parents that have children with dysfunctions can have a problem applying alternative and non-violent discipline since these children can have a problem understanding them, communicating, or hearing them [50].Moreover, increased pressure of caring for a child with dysfunctions, additional parenting tasks, and behavior management difficulties can, in turn, lead to an increase in parental stress, and the likelihood of believing that CP will be an efficient short-or long-term disciplining method [51].
The results from the present study also indicate that the attitudes of parents of children with reported dysfunctions towards the use of CP are positive.When each group of dysfunctions was analyzed we found that parents of children with psychological and academic dysfunctions have positive attitudes towards CP, while the attitudes of parents of children with developmental dysfunctions did not differ from the attitudes of parents with children without dysfunctions.In a recent study in Taiwan [27], it was shown that isolated parental attitudes did not predict the negative outcomes in children.However, when the positive attitudes are combined with the actual use of CP, then attitudes can significantly predict later dysfunctions in children such as depression and violent behavior [27].
The final aspects of the parental discipline that we analyzed are the parental practices.From the results presented in our study, it can be concluded that the children with any reported problem are significantly more often subjected to all forms of disciplining.This is not a surprising result, mainly because children with difficulties usually require more parental attention, support, and guidance to achieve their full potential.However, it should be noted that the children with dysfunctions are not only disciplined more by non-violent practices, but they are also more often exposed to almost all types of violent parental practices.In the past, the researchers constantly reported that children with disabilities and behavioral disturbances are more frequently exposed to violent practices (physical, psychological), and other forms of abuse and neglect [16][17][18][19][20][52][53][54].According to the authors, this is mainly due to their common cognitive immaturity and social invisibility [1].There is an increasing body of evidence that CP and other forms of harsh punishment practices and abuse can alter brain function [52,55].It needs to be highlighted that these children are already more vulnerable to all types of adverse external factors, and even with sufficient support often struggle to reach their full potential and adequately follow their peers.
When parental expectations of CP outcomes, parental attitudes, and parental practices were analyzed as predictors of the reported child dysfunctions severity among the participants, significant effects were found.The significant predictors for the reported developmental dysfunctions in children were the parental positive expectations of CP (i.e., positive outcomes) as well as the parental use of physical assault.The significant predictors of academic dysfunctions in children were psychological aggression, minor physical assault and neglect.The significant predictors for psychological-emotional dysfunctions in children were all forms of physical abuse (minor to extreme).Considering the fact that this is a cross-sectional study, the associations between the examined variables can be bidirectional.The use of CP and other harsh disciplining practices can lead to many negative mental health outcomes for a child such as internalizing (depression, NSSI, suicidal behavior) [12,13] and externalizing problems (increased impulsivity, temper outbursts, and aggression) [11].This is why parents need to bear in mind the fact that not only these violent measures are not going to resolve child's dysfunctions, but they can cause and/or worsen them.On the other hand, the challenging behaviors that are often expressed by children with dysfunctions can lead to a decrease in parental capacity to handle their demanding parental role and to deal with stress, which can worsen their unfavorable discipline practices.These factors can form a virtuous circle that can be dangerous for the whole family.
The main limitation of this study is its cross-sectional structure and its inability to form a causal relationship between the measured variables.The use of self-report and the retrospective nature of some parts of the study instrument may have caused a recall bias among the participants.In addition, this study was based on parental reports of child dysfunctions, which could have led to a false negative and false positive identification of certain dysfunctions in their children.The participants were approached by the interviewer in their household and the interview was conducted face-to-face.Some questions could have been sensitive and possibly incriminating for certain parents, which is why they might have been reluctant to provide truthful answers.

Conclusion
Based on the findings from this research, there is evidence that parents of children with certain dysfunctions have positive attitudes toward violent punishment, as well as positive expectations of CP.In addition, the children with dysfunctions are more often subjected to CP and other forms of violent disciplining.Having all the results in mind, we can assume that children with health-related and school-related issues might be at potential risk of further maltreatment.More studies are needed to assess the exact additive effects and interaction between these factors, as well as causal relationships between the factors related to children and parental attitudes, norms and practices.
This study highlights the need for close cooperation between the researchers and policymakers that need to provide comprehensive systems of support and education which can help in the prevention of child maltreatment.As seen from this study, these systems need to be especially directed towards parents and families that care for a child with dysfunctions due to their increased vulnerability.Since parental opinions and attitudes are a large piece of the puzzle related to parental practices, the education of parents on "positive parenting" and adequate disciplining measures should be at the forefront of preventive interventions.

Fig. 1
Fig.1Summary of the relationships between the expected outcomes of CP, attitudes towards CP and disciplinary practices and the different types of children's difficulties expressed via beta (regression) coefficient.Note that only significant findings were shown in the figure

Table 1
The socio-demographic data of parental sample

Table 5
Disciplinary practices in families with children with any reported problemNote that the CTSPC scale measured the frequency of practices in the "past year" (12 months before the moment of research), and "in lifetime" (by marking the response "not in the past year, but it happened before").*p < 0.05; **p < 0.001