Assessment, frequency, and reasons for child-to-parent violence in a sample of Italian adolescents

Abstract

Scientific literature shows that child-to-parent violence (CPV) is a worrying type of family violence present in different sociocultural contexts. However, the frequency of CPV is still unclear, especially in countries such as Italy, where the research is practically nonexistent and which also do not have validated CPV assessment tools adapted to their population. The aims were to analyze the psychometric properties of the Italian version of the Child-to-Parent Violence Questionnaire (CPV-Q) and to examine the frequency and reasons for CPV in a sample of Italian adolescents. 562 adolescents (54.4% girls) aged between 13 and 18 years from educational centers in two regions of Italy participated. A population-based study was designed using cross-sectional surveys in which participants completed the CPV-Q. The original structure of the CPV-Q was validated with exceptional results, obtaining outstanding fit indices. The instrument demonstrated robust reliability in all its scales, robust validity in almost all scales, and total invariance according to the adolescent’s sex. High rates of CPV were found, with daughters showing more psychological (toward father), physical (toward mother), and control-domain (toward both parents) violent behaviors than sons, and mothers experiencing all types of CPV at higher rates than fathers. Daughters exercise CPV for more reactive reasons than sons. The study suggests a high prevalence of CPV among Italian adolescents, so urgent research resources are needed in the country. The CPV-Q can be used as an effective tool to identify and assess CPV in this population.

Introduction

One of the greatest problems today is the violence from children to parents, better known as child-to-parent violence (hereinafter CPV). This type of family violence causes significant deterioration in family dynamics and well-being, with serious repercussions also at the individual and social levels of parents^1–3 and children^4–7. Despite the growing interest in research on this type of violence in the last decade, it remains one of the least studied forms of family violence. Concern among professionals and researchers has grown due to the notable increase in these behaviors in recent years⁸, a situation even more alarming considering that a large part of the cases goes undetected but remains hidden in the privacy of the home^9,10.

Efforts to define CPV have been ongoing since its first mention in the literature¹¹. The first documented definition described the phenomenon as “actual physical assaults or verbal and nonverbal threats of physical harm"¹², p. 1288, whereas today, one of the most widely recognized definitions describes it as “any act by a child intended to cause physical, psychological, or financial harm to gain power and control over a parent"¹³, p. 3. In addition, a series of elements have been incorporated into the conceptualization of the phenomenon. Thus, acts of violence are carried out intentionally, consciously, and repeatedly over time¹⁴ in the context of habitual family cohabitation¹⁵ and are directed at both biological parents and other caregivers who take their place^15,16. Likewise, the fundamental role played by the control and domain of parental figures in this type of violence is highlighted^2,13,17. Consequently, isolated episodes of violence, those occurring in states of diminished consciousness or withdrawal syndromes, and parricide with no history of aggression would be excluded¹⁴.

Following authors such as Cottrell¹³ or Howard and Rottem¹⁷, physical CPV includes behaviors that produce physically observable harm to parents and are manifested in acts such as pushing, spitting, kicking, and hitting. Psychological CPV includes behaviors intended to cause emotional harm, such as degrading, humiliating, insulting, belittling, or threatening to run away from home. Financial CPV refers to behaviors that restrict the parent’s earning and saving possibilities, such as stealing money, selling parental belongings, or destroying household property. Finally, control and domain behaviors are aimed at limiting the parent’s freedom of action and decision and include actions such as coercing, extorting, frightening, impeding speech, imposing issues, psychological control, or blaming.

The literature shows great variability in the estimated prevalence rates in the community population¹⁸, which limits our understanding of the real extent of this problem. For example, in Anglo-Saxon, Latin American, and European countries (see Table 1), physical CPV among adolescents ranges between 0.9 and 25%, psychological between 3.1 and 95.3%, financial between 9.3 and 50.9%, and control and domain behaviors between 19.4 and 66.6%^4,19–31. These variations are mainly attributed to the lack of consistency in both the definition of CPV¹⁵ and the criteria used to determine its presence³². Most studies have applied the zero-tolerance criterion and the technical criterion. The first refers to occasional violence so that it establishes the presence of CPV when the behavior is exercised on at least one occasion. The second criterion refers to repeated violence, a criterion more consistent with the current definition of CPV¹⁴.

Table 1.

Prevalence of occasional and reiterated child-to-parent violence by adolescents in the community population.

Country	Children’s age (years)	Physical		Psychological		Financial		Control-Domain
		Occasional	Reiterated	Occasional	Reiterated	Occasional	Reiterated	Occasional	Reiterated
USA19,20,25,28,31	9–18	2.0–25.0	1.5	54.7–80.0		35.8–50.9
Canada29,30	15–16	9.5–13.7		53.5–65.9
Mexico23	14–18	5.1–7.3	0.9–1.7	71.2–89.6	3.1–4.1
Ecuador21	12–18	5.1–6.8	1.8–2.3	41.3–47.5	19.6–25.2	37.9–42.8	15.1–17.5	53.0–62.8	30.5–35.5
Chile27	12–18		2.5–5.4		25.2–28.8		9.3–13.2		19.4–25.4
Germany4	13–19	5.5		45.3
Spain22,24,26	12–18	5.0–8.1	1.9–2.5	43.2–95.3	5.8–24.0	31.4–38.0	9.9–15.9	58.5–66.6	28.3–36.0

In the community adolescent population, the occasional prevalence of physical CPV ranges from 2 to 25% and psychological from 41.3 to 95.3%^{4,19,21–26,28–31}, while repeated physical CPV ranges from 0.9 to 5.4% and psychological from 3.1 to 28.8%^{4,19,21–24,26,27}. Financial CPV and, above all, control-domain behaviors have been investigated by a smaller number of studies. Even so, the available data also show high percentages of these types of CPV in different countries (see Table 1). Specifically, financial CPV ranges from 31.4 to 50.9%, with percentages of repeated violence of 9.3–17.5%^21,24,28, and control-domain behaviors occur between 53 and 66.6% of cases, with percentages of repeated violence of 19.4–36%^21,24,27.

Regarding sex differences in these percentages, Contreras et al.²⁴ found that sons between 12 and 18 years of age exercised more financial violence toward fathers, while daughters exercised more psychological violence toward fathers and mothers, as in another Spanish sample²². Psychological violence is also more frequent from daughters to mothers in a USA study²⁸ and in Latin American samples^21,23. Likewise, more girls than boys exercised control and domain behaviors toward mothers^21,24 and fathers²¹. No differences are found in the proportion of physical CPV between daughters and sons. Regarding differences according to parents’ sex, studies from different countries consistently show significantly higher proportions of CPV toward mothers than fathers^23,24,27,28. Only in one study did fathers experience more physical CPV than mothers²⁷.

Data from studies conducted in different countries also vary, even within the same country. For example, Table 1 shows that occasional physical CPV reaches 25% in the USA studies, while in Germany, it does not reach 6%, and psychological CPV in Mexico occurs repeatedly in 4% of cases, while in Chile, it is close to 30%. In Spain, 90.7% of boys have exercised psychological CPV on occasion towards their mothers in the study by Calvete and Orue²², while in the study by Contreras et al.²⁴, this percentage is 46.7%. It may be that the pattern of CPV differs depending on the sociocultural factors present in each country, so cross-cultural research, practically nonexistent³³, is needed to clarify this question. However, it has been clarified in the literature that the lack of consistency in the definition and measurement of CPV has led to variations in how this phenomenon is assessed. Specifically, it has been pointed out that the use of various assessment instruments to measure CPV could be the main reason for the disparity in CPV figures^15,34.

Currently, there are more than ten instruments to assess CPV^15,34. Ibabe¹⁵ in her systematic review concludes that the most widely used instrument has been the Conflicts Tactics Scale (CTS³⁵), while the most prominent for application in research and clinical settings according to COSMIN criteria is the Child-to-parent Violence Questionnaire (CPV-Q³⁶), the Adolescent Child-to-Parent Aggression Questionnaire (CPAQ³⁷), and the Abusive Behavior by Children-Indices (ABC-I⁷). Burgos-Benavides et al.³⁴, in their meta-analysis, identify the CPV-Q and the CPAQ as the most promising options due to their high levels of reliability and their wide application in diverse samples from different countries. In addition, both instruments assess the reasons for CPV, providing a more comprehensive assessment of the motivational mechanisms underlying this type of violence.

Regarding the reasons for CPV, there are fundamentally two types: instrumental and reactive^22,36. Instrumental reasons involve the use of violence for personal gain, while reactive reasons are described as the response to a real or perceived threat or provocation³⁸. The few studies that have analyzed the reasons for CPV find that both types tend to coexist intraindividual^2,39,40, as well as that instrumental ones are more frequent in adolescent boys^21,23,24 and reactive ones in girls^21–24,27.

The literature shows that research on this topic is gaining impulse in Latin American countries³², although most of the studies are concentrated in some Anglo-Saxon and European countries¹⁸. Despite this, the review conducted in the present study shows that the frequency of CPV and the reasons for its occurrence are still unclear, so first, comprehensive CPV assessment tools that have proven their validity would be necessary. This situation is even more worrying in certain countries such as Italy, where studies on CPV are practically nonexistent and which also do not have psychometric instruments adapted to their population. Specifically, only three studies have been found that analyze CPV in Italy. A qualitative study⁴¹ conducted in Bologna, Emilia-Romagna, using snowball sampling, involved 27 professionals with experience in CPV. These participants, from diverse sectors such as child neuropsychiatry, youth services, and juvenile courts, provided a well-rounded view of the situation. Although the study was limited in scope, it aimed to spark reflection and encourage further research to develop appropriate intervention strategies to mitigate the familial and social impacts of adolescent violence towards parents. The findings of this study highlight the transversality of CPV aimed at causing harm of a physical, psychological, and/or financial nature and the dynamics of power and control¹³. Violence is directed at both parents, though mothers are more frequently targeted. Gender differences also emerge: Boys tend to exhibit more physical aggression, while girls engage in more psychological violence. Maggiolini et al.⁴², examined 528 cases at the Juvenile Court of Milan from 2019 and revealed that 14% of behaviors involved CPV, indicating a significant issue. The study emphasized that CPV should be viewed within the broader dynamics of family roles in today’s sociocultural context, rather than as a result of past abuse or psychopathological conditions. Similar conclusions are found in the study by Abazia⁴³.

Given that CPV is a global issue with serious consequences, it is urgent to provide researchers with validated CPV assessment tools in the Italian population that allow them to know the magnitude of CPV and the factors involved in its development. It is crucial to prevent the occurrence of this type of violence through early detection and to implementation of effective evidence-based interventions. Among the instruments highlighted in previous reviews^15,34, in the present study, we opted for the CPV-Q³⁶ because it is the instrument most consistent with the current definition of CPV by assessing a wide variety of behaviors of psychological, physical, and financial CPV, as well as manifestations of control and domain over parents¹³, a dimension that is not incorporated in other instruments. In addition, CPV-Q also assesses the reasons for carrying out CPV and has been validated in several sociocultural contexts, demonstrating good psychometric properties in Spanish³⁶, Chilean²⁷, Ecuadorian²¹, and Chinese⁴⁴ adolescents, which a priori supports its usefulness for assessing CPV also in Italian adolescents.

The present study

The purpose of this study is to provide a psychometric instrument to assess CPV in the Italian population and to know the frequency of this type of violence and the reasons for carrying it out in a sample of Italian adolescents. Four specific objectives were set:

1. To examine the factor structure, reliability, and validity of the Italian version of the Child-to-Parent Violence Questionnaire (CPV-Q adolescents version; ³⁶) in a sample of Italian adolescents; H1. The four-factor structure of the violence scale (psychological, physical, financial violence, and control-domain behaviors) and the two-factor structure of the reasons subscale (instrumental and reactive reasons) of the CPV-Q will be confirmed, demonstrating adequate reliability, convergent, and discriminant validity^21,27,36,44.

2. To explore the measurement invariance of the CPV and its reasons based on the participant’s sex; H2. The CPV-Q concerning the mother is expected to be invariant between boys and girls²¹.

3. To analyze the frequency of the different types of CPV toward fathers and mothers and the differences based on the children and parents’ sex; H3. Daughters will exert more psychological violence^21–24,28 and control-domain behaviors^21,24 toward mothers and fathers. H4. CPV will be more frequent toward mothers than fathers^23,24,27,28.

4. To examine the reasons for violence towards fathers and mothers and the differences based on the children and parents’ sex; H5. Instrumental reasons will be more frequent in boys^21,23,24 and reactive in girls^21–24,27.

Methods

Participants

The initial sample consisted of 620 participants, of whom those who indicated that they habitually lived with at least one parent (92.6%) and met the age criterion (98.1%) were included. The final sample was composed of 562 adolescents (45.6% boys and 54.4% girls) of Italian nationality (88.1%) aged between 13 and 18 years (M_age = 15.82, SD = 1.59) coming from schools located in 3 Italian provinces, most of them from Emilia-Romagna (64.1%) and Sicilia (35.7%). At the time of data collection, the distribution of the sample by the five compulsory academic years in Italy was: first (17.4%), second (22.2%), third (29.9%), fourth (17.4%), and fifth (13.0%). The marital status of parents (98.9% biological; 1.1% adoptive) was predominantly married (73.9%), with the remainder being divorced or separated (15.6%; shared custody (31.3%), exclusive of the father (10%), exclusive of the mother (58.8%)), unmarried but cohabiting (7.5%), widowed (2.3%), or never having lived together (0.7%). Monthly household income ranges between 1,000€-2,000€ (39.1%), between 2,000€-3,000€ (27.6%), more than 3,000€ (20.4%), and less than 1,000€ (4.6%).

Instruments

Brief sociodemographic ad hoc questionnaire: It comprises a series of questions aimed at gathering sociodemographic information about the participants (sex, age, nationality, place of residence, academic year) and their family unit (household composition, parents’ marital status, which parent has custody, household income).

Child-to-Parent Violence Questionnaire, adolescents version (CPV-Q,³⁶, see Supplementary Information File, “Appendix A”). It assesses the frequency of a wide variety of violent behaviors toward parents through 14 parallel items (violence towards father and mother separately) divided into four dimensions (psychological, physical, financial violence and control-domain behaviors) answered on a 5-point Likert scale: 0 = never (has never occurred); 1 = rarely (has occurred once); 2 = sometimes (has occurred two or three times); 3 = many times (has occurred four or five times); and 4 = very often (has occurred six times or more). The temporal frame refers to the frequency with which the violent behaviors have occurred during the last year. The second part assesses the reasons for carrying out violent behaviors through eight parallel items (reasons for violence towards father and mother separately) distributed in two dimensions (instrumental and reactive reasons) answered on a 4-point Likert scale: 0 = never; 1 = sometimes; 2 = almost always; and 3 = always.

Procedure¹

The study was conducted according to the Declaration of Helsinki and was approved by the Institutional Review Board of the University of Jaén (protocol number CEIH 270215-1). For the adaptation and linguistic translation of the CPV-Q³⁶, native specialists analyzed the content information of the construct, the linguistics, and the grammar of the questionnaire items and instructions⁴⁵. The research has been possible because of the collaboration with two local school authorities: one of them has its competencies in the provinces of Forlì-Cesena and Rimini, located in the Emilia-Romagna region (Northern Italy); the other one in the province of Messina, located in the Sicily region (Southern Italy). There are two reasons for having chosen these territories: the first one is the methodological general need to differentiate the geographical origins of the respondents; the second one is the collaboration network established by the two Italian researchers that has provided easier access to those areas in particular. To gain the cooperation of all teachers of middle and high schools based in these three provinces (Forlì-Cesena, Rimini, and Messina), the managers of these two schools’ local authorities sent them an e-mail request with an attached information letter prepared by the researchers. The parental informed consent for the participation of their minor children was obtained. The researchers were authorized to go to the classrooms to explain the objectives of the study to the students and how to complete the questionnaire, which was administered online through the Google Forms platform (University of Jaén license). The occasion was also used to increase participation, while emphasizing its voluntary nature, encouraging them to respond sincerely to questions as there were no right or wrong answers and to assure them both of the confidentiality of the information collected and of the guarantee of privacy due to the recording of data being in an aggregated and anonymized form.

Data analysis

Statistical analyses were performed in R software version 4.3.2⁴⁶. First, the asymmetry, kurtosis, and mean scores for each of the items were analyzed. Multivariate normality was analyzed for each of the scales. Once the assumptions for the confirmatory factor analysis (CFA) were confirmed, a CFA was performed for the CPV-Q scales (violence toward fathers and mothers) and the reasons subscales (reasons for violence toward fathers and mothers). The estimation method used was Diagonally Weighted Least Squares (DWLS) due to the multivariate non-normality of the data and the categorical/ordinal nature of the scales⁴⁷. Conventional goodness-of-fit indices based on a measure of the comparative fit to the null model of independence were used to evaluate the fit of the model. Specifically, the Root Mean Square Error of Approximation (RMSEA), the Comparative Fit Index (CFI), and the Standardized Root Mean Square Residual (SRMR). The model fit is considered adequate with RMSEA parameters less than or equal to 0.06^48,49, CFI greater than or equal to 0.95^49,50, and SRMR less than or equal to 0.08⁴⁹.

Reliability was assessed using Ordinal Cronbach’s Alpha and Ordinal McDonald’s Omega. Although Cronbach’s Alpha is the most widely used coefficient to measure the reliability of scales in psychology, it is recommended to use McDonald’s Omega, which is more robust and accurate in the context of a CFA⁵¹. Both indices are considered acceptable at ≥ 0.70^52,53, at ≥ 0.65⁵¹, or at ≥ 0.60^54,55. The Composite Reliability Coefficient (CR) is also presented, which is similar to omega but is a particularly useful indicator for explaining latent constructs. The Average Variance Extracted (AVE) was used to evaluate the convergent validity that provides the average amount of variance of the items explained by each factor. Some authors suggest that the AVE should explain 50% of the variance⁵⁶, although authors such as Moral de la Rubia suggest that an AVE of 44% and 37% could be acceptable for scales of three and four indicators, respectively⁵⁷. Discriminant validity was assessed using the Heterotrait-Monotrait Relationship Technique (HTMT2). This type of validity consists of assessing the relationship between correlations between constructs and correlations within constructs. Values below 0.75 are usually considered an adequate criterion of discriminant validity^56,58,59.

Measurement invariance is a fundamental property for analyzing the equivalence of scales between groups, especially when making comparisons between groups⁵¹. In this study, a sex invariance analysis was performed for each of the CPV-Q scales (fathers and mothers) and the reasons subscales (fathers and mothers) through the Lavaan package of R⁶⁰. The procedure consisted of checking the configural level, and once it was verified that the factor structure was the same in both groups, metric invariance was checked to verify that the factor loadings were the same for sons and daughters. Subsequently, scalar invariance was calculated to determine the equivalence between the intercepts. Finally, we tested whether the scales showed strict invariance, i.e., whether the residuals were equivalent in both groups^61,62. The baseline criterion for moving to the next level of invariance measure was that the increase in the root mean square of the approximation residuals (RMSEA) was no more than 0.015 and the increase in the comparative fit index (CFI) was no more than 0.010⁶³.

To analyze the prevalence of CPV, we calculated the percentages of adolescents who had engaged in violence towards their father and mother using two criteria: (1) the presence of any violent behavior (zero tolerance criterion: a score of 1 or more 'has occurred once’) and (2) the presence of repeated violent behavior (reiterated criterion: a score of 2 or more 'has occurred sometimes, two or three times’). The frequency of the CPV-Q factors of the violence scale (psychological, physical, financial, and control/domain) was assessed. Next, we analyzed sex-based differences in CPV types using Pearson’s Chi-Square test for independent samples to compare children’s sex and McNemar’s Chi-Square test for related samples to compare parents’ sex. The effect size was measured using the Phi coefficient.

Finally, the mean scores of the CPV-Q factors of the reasons subscale (instrumental and reactive) were evaluated, and we analyzed sex-based differences through t-tests for independent samples to compare children’s sex and through t-tests for related samples to compare parents’ sex. The effect size was assessed with Cohen’s d.

Results

Descriptive statistics

Table 2 presents the mean scores, skewness and kurtosis of the items on the scales. Multivariate normality analysis revealed that the data did not meet the assumption of normality in the CPV-Q scales (toward fathers: Z_kurtosis = 241.298, p < 0.001; toward mothers (Z_kurtosis = 248.561, p < 0.001) and the reasons subscales (toward father: Z_kurtosis = 52.070, p < 0.001; toward mothers: Z_kurtosis = 51.521, p < 0.001).

Table 2.

Descriptive statistics of the child-to-parent violence questionnaire adolescents version.

Scale	Factor	Item	Father (n1 = 528)				Mother (n2 = 553)
			M	SD	SK	KT	M	SD	SK	KT
Violence	Psychological	1	0.22	0.64	3.51	13.43	0.31	0.75	2.94	9.18
		2	0.75	1.03	1.54	1.85	0.88	1.13	1.34	1.07
		3	0.63	0.96	1.82	3.17	0.67	1.02	1.70	2.39
		4	0.22	0.67	3.62	13.95	0.28	0.73	3.08	9.71
	Physical	8	0.08	0.34	4.64	23.97	0.16	0.51	3.90	17.06
		10	0.04	0.20	5.74	35.36	0.06	0.31	6.54	49.70
		11	0.08	0.38	5.84	41.11	0.07	0.37	6.86	57.81
	Financial	6	0.20	0.62	3.94	17.72	0.23	0.65	3.60	14.71
		7	0.07	0.39	7.23	59.56	0.08	0.38	6.17	44.69
		12	0.29	0.70	3.01	10.00	0.35	0.79	2.71	7.63
	Control/Domain	5	1.32	1.24	0.52	-0.78	1.30	1.25	0.60	-0.62
		9	1.10	1.25	1.02	0.03	1.31	1.30	0.73	-0.58
		13	0.18	0.58	4.00	17.65	0.24	0.67	3.54	13.75
		14	0.61	0.92	1.67	2.64	0.78	1.03	1.39	1.45
Reasons	Instrumental	1	0.69	0.93	1.27	0.64	0.76	0.95	1.12	0.28
		2	0.40	0.75	1.91	2.97	0.41	0.76	1.91	2.98
		3	0.56	0.81	1.42	1.38	0.63	0.87	1.31	0.84
		4	0.42	0.69	1.75	2.88	0.55	0.80	1.50	1.75
		5	0.38	0.65	1.91	3.99	0.49	0.72	1.59	2.46
	Reactive	6	0.80	0.96	1.03	0.05	0.86	0.98	0.88	-0.31
		7	0.24	0.59	2.93	8.90	0.23	0.55	2.82	8.67
		8	0.45	0.73	1.81	3.18	0.49	0.78	1.69	2.36

n = participants who during the last year have had regular contact with father (¹) and mother (²); M = mean; SD = standard error; SK = skewness; KT = kurtosis.

Confirmatory factor analysis

Table 3 presents the factor loadings of the CPV-Q scale and the reasons subscale. All items of the violence scale (except item 5 used to explain control and domain behaviors) and the reasons subscale present factor loadings above the minimum acceptable level (> 0.50).

Table 3.

Factor loading of the child-to-parent violence questionnaire adolescents version.

Items	Father (n1 = 496)				Mother (n2 = 530)
Violence	Psychological	Physical	Financial	C/D	Psychological	Physical	Financial	C/D
1	.79				.80
2	.88 (.06)				.85 (.04)
3	.86 (.06)				.84 (.04)
4	.79 (.07)				.77 (.05)
8		.91				.88
10		.80 (.09)				.87 (.05)
11		.82 (.07)				.90 (.05)
6			.73				.69
7			.57 (.13)				.59 (.08)
12			.57 (.11)				.72 (.08)
5				.25				.20
9				.59 (.52)				.63 (.42)
13				.81 (.75)				.83 (.55)
14				.64 (.58)				.69 (.46)

Reasons	Instrumental	Reactive			Instrumental	Reactive
1	.77				.80
2	.84 (.05)				.82 (.04)
3	.83 (.05)				.86 (.05)
4	.79 (.05)				.80 (.04)
5	.67 (.05)				.72 (.04)	.74
6		.73				.84 (.07)
7		.88 (.08)
8		.77 (.07)				.79 (.07)

n = participants who during the last year have had regular contact with father (¹) and mother (²); C/D = control/domain; The parameters in parentheses refer to the standard error.

The CFA shows that the factor structure of the CPV-Q explains the violence toward parents through a four-factor model. Specifically, the results showed an excellent model fit on both the father scale (² = 64.43 (71), p = 0.696, CFI = 0.99, TLI = 0.99, RMSEA = 0.00 [0.00–0.02], SRMR = 0.06, = 0.81, = 0.84, G6 = 0.82) and the mother scale (² = 86.60 (71), p = 0.100, CFI = 0.99, TLI = 0.99, RMSEA = 0.02 [0.00–0.03], and SRMR = 0.06, = 0.83, = 0.86, G6 = 0.85). The reasons for this violence are explained through a two-factor model, toward fathers (² = 113.12 (19), p < 0.001, CFI = 0.98, TLI = 0.96, RMSEA = 0.10 [0.08–0.12], SRMR = 0.09, = 0.80, = 0.86, G6 = 0.81) and mothers (² = 111.97 (19), p < 0.001, CFI = 0.98, TLI = 0.97, RMSEA = 0.09 [0.08–0.11], SRMR = 0.08, = 0.83, = 0.86, G6 = 0.85).

Reliability and convergent and discriminant validity

Table 4 shows the reliability, convergent, and discriminant validity of the CPV-Q toward the father and mother (violence scale and reasons subscale).

Table 4.

Reliability and convergent and discriminant validity of the child-to-parent violence questionnaire adolescents version.

	Father (n1 = 496)									Mother (n2 = 530)
	α	αo	Ω	Ω0	CR	AVE	HTMT2			α	αo	Ω	Ω0	CR	AVE	HTMT2
							Psychological	Physical	Financial							Psychological	Physical	Financial
Violence
Psychological	.81	.89	.84	.89	.84	.69	–			.82	.88	.85	.90	.83	.66	–
Physical	.65	-	.66	-	.71	.72	.64	–		.73	-	74	-	.78	.78	.71	–
Financial	.49	.67	.50	.69	.44	.40	.51	.48	–	.55	.72	.55	.73	.55	.45	.55	.30	–
Control/Domain	.53	.65	.58	.71	.48	.37	.53	.41	.58	.58	.67	.61	.71	.53	.40	.55	.37	.64
Reasons							Instrumental									Instrumental
Instrumental	.80	.87	.86	.92	.85	.61	–			.82	.88	.87	.93	.88	.64	-
Reactive	.70	.82	.73	.83	.75	.64	.48			.70	.82	.73	.83	.74	.63	.55

n = participants who during the last year have had regular contact with father (¹) and mother (²); α = Cronbach’ Alpha; α_o = Ordinal Cronbach’ Alpha; Ω = McDonald’ Omega; Ω_o = Ordinal McDonald’ Omega; CR = composite reliability; AVE = average variance extracted; HTMT2 = heterotrait-monotrait ratio.

The highest scores in the reliability coefficients correspond to the factors of psychological violence (0.88–0.90), followed by physical violence (0.65–0.74) financial violence (0.67–0.73) and control-domain behaviors (0.65–0.71). The reliability of the reasons subscale was adequate for both instrumental (0.87–0.93) and reactive factors (0.82–0.83).

The AVE of the violence scale indicated that the items of psychological and physical violence factors converge in a high proportion (> 50%) of common variance to explain each of the factors. The items of control/domain behaviors (4 indicators ≥ 37%) and financial violence toward the mother (3 indicators ≥ 44%) converge in an acceptable proportion. Only the financial violence toward the father factor (0.40) does not reach the recommended threshold. Regarding the reasons for CPV, AVE shows that the items of both factors converge in a high proportion (> 50%) to explain through instrumental and reactive reasons the children’s motives for exercising CPV.

The HTMT2 shows that in all cases (violence scale and reasons subscale toward fathers and mothers) the correlations between the constructs were less than 0.75, indicating good discriminant validity between them.

Factorial invariance models

The invariance analyses of the CPV-Q (violence scale and reasons subscale toward fathers and mothers) were performed according to the sex of the participants (see Table 5).

Table 5.

Invariance measures of the child-to-parent violence questionnaire adolescents version according to children’s sex.

	Models	x2/df	RMSEA (90% CI)	ΔRMSEA	SRMR	ΔSRMR	CFI	ΔCFI	TLI
Father1	Violence
	CF	68.659/142	.000 (.000–.000)		.061		.999		.999
	MT	119.313/152	.000 (.000–.000)	.000	.072	.011	.999	.000	.999
	SC	129.014/162	.000 (.000–.000)	.000	.074	.002	.999	.000	.999
	ST	146.051/176	.000 (.000–.000)	.000	.093	.019	.999	.000	.999
	Reasons
	CF	55.058/38	.043 (.011–.066)		.072		.978		.968
	MT	63.800/44	.043 (.015–.064)	.000	.077	.005	.974	− .004	.967
	SC	67.064/50	.037 (.000–.059)	− .006	.079	.002	.978	.008	.975
	ST	75.756/58	.035 (.000–.056)	− .002	.088	.009	.977	− .001	.978
Mother2	Violence
	CF	94.049/142	.000 (.000–.000)		.068		.999		.999
	MT	129.203/152	.000 (.000–.012)	.000	.082	.014	.999	.000	.999
	SC	138.652/162	.000 (.000–.012)	.000	.084	.002	.999	.000	.999
	ST	160.228/176	.000 (.000–.019)	.000	.102	.018	.999	.000	.999
	Reasons
	CF	46.453/38	.029 (.000–.055)		.067		.992		.988
	MT	50.829/44	.024 (.000–.050)	− .005	.070	.003	.993	.001	.992
	SC	53.733/50	.017 (.000–.044)	− .007	.071	.001	.996	.003	.996
	ST	60.811/58	.014 (.000–.041)	− .003	.080	.009	.997	.001	.997

Participants who during the last year have had regular contact with father (¹n = 496) and mother (² n = 530); CF = configural; MT = metric; SC = scalar; ST = strict.

Total invariance was found between sons and daughters. More specifically, it was found configural invariance in the four-factor violence scale and the two-factor reasons subscale. The factor loadings are also equivalent between both groups (metric invariance), item intercepts are the same for both groups (scalar invariance), and the measurement errors were equal between the two groups (strict invariance).

Frequency of child-to-parent violence types

Table 6 shows high percentages of CPV in the sample. Specifically, between 53.8 and 64.10% of the adolescents have exercised any psychological violence behavior at least once, and between 17.9 and 31.9% have done so repeatedly. Physical violence ranged between 9.7 and 14.1%, with percentages of repeated violence between 2.4 and 5.9%. Financial violence was reported occasionally by 27.6–35% of adolescents, with 8.4–11.9% engaging in this CPV type repeatedly. Control-domain behaviors occur in 78.9–89.1% of cases occasionally and in 50.4–69.9% of cases repeatedly.

Table 6.

Frequency of child-to-parent violence toward father and mother and differences according to children’s and parent’s sex.

Violence	Father (n 1 = 528)				Mother (n 2 = 553)				Both girls and boys (n 3 = 519)
	Sons % (n)	Daughters % (n)	2		Sons % (n)	Daughters % (n)	2		Fathers % (n)	Mothers % (n)	2
At least 1 occasion
Psy	53.80 (129)	60.70 (173)	2.57	.07	63.50 (158)	64.10 (195)	0.03	.01	57.80 (298)	64.90 (335)	261.15***	.61
Phy	9.70 (23)	11.60 (33)	0.49	.03	12.90 (32)	14.10 (43)	0.18	.02	10.09 (56)	14.00 (72)	194.09*	.61
Fin	27.60 (66)	29.90 (85)	0.34	.02	31.90 (79)	35.00 (106)	0.60	.03	29.40 (151)	33.50 (172)	286.45**	.75
C/D	78.90 (187)	88.10 (251)	8.05**	.12	87.00 (215)	89.10 (271)	0.58	.03	84.20 (432)	88.50 (454)	250.28***	.70
Reiterated violence
Psy	17.90 (43)	26.70 (76)	5.69*	.10	25.00 (62)	31.90 (97)	3.18	.08	22.70 (117)	29.30 (151)	183.85***	.60
Phy	2.50 (06)	4.20 (12)	1.11	.05	2.40 (06)	5.90 (18)	4.03*	.08	3.50 (18)	4.70 (24)	108.51	.46
Fin	8.40 (20)	11.60 (33)	1.51	.05	10.90 (27)	11.90 (36)	0.13	.02	10.30 (53)	11.30 (58)	272.64	.73
C/D	50.40 (119)	68.70 (195)	17.92***	.19	58.80 (144)	69.90 (211)	7.31**	.12	61.10 (312)	65.00 (332)	238.98*	.68

n = participants who during the last year have had regular contact with father (¹), mother (²), and with both father and mother (³); Psy = Psychological; Phy = physical; Fin = financial; C/D = control/domain.

*p < 0.05, **p < 0.01, ***p < 0.001.

In addition, there are statistically significant differences in the frequency of CPV as a function of the sex of both children and parents. Specifically, compared to sons, daughters exert more psychological violence toward fathers, physical violence toward mothers, and control-domain behaviors toward both parents. Regarding parents’ sex, all CPV types are more frequent toward mothers than fathers when carried out occasionally. Repeated violence is also more frequent towards mothers. Specifically, more mothers than fathers repeatedly experience psychological violence as well as control and domain behaviors by their children.

Reasons for child-to-parent violence

Table 7 shows that the most frequent reason reported for CPV was “because of your own temper”, followed by “to be able to come home later when you went out at night”, “to get your father/mother to buy you something that you wanted”, and “in response to a previous verbal aggression from your father/mother (e.g., insult)”.

Table 7.

Reasons for child-to-parent violence toward father and mother and differences according to children’s and parent’s sex.

Reasons	Father (n1 = 528)				Mother (n2 = 553)				Both girls and boys (n3 = 519)
	Sons M (SD)	Daughters M (SD)	t	d	Sons M (SD)	Daughters M (SD)	t	d	Fathers M (SD)	Mothers M (SD)	t	d
Instrumental	2.76 (3.09)	2.35 (2.85)	1.51	0.14	3.01 (3.13)	2.77 (3.24)	0.87	0.07	2.55 (2.99)	2.91 (3.21)	− 4.87***	0.22
1	0.78 (0.98)	0.66 (0.92)	1.43	0.13	0.83 (0.96)	0.74 (0.97)	1.14	0.09	0.72 (0.95)	0.79 (0.97)	− 3.19**	0.14
2	0.47 (0.81)	0.38 (0.74)	1.27	0.12	0.44 (0.80)	0.40 (0.74)	0.65	0.05	0.42 (0.77)	0.42 (0.76)	0.19	0.01
3	0.61 (0.82)	0.56 (0.83)	0.72	0.06	0.67 (0.88)	0.61 (0.87)	0.78	0.07	0.58 (0.82)	0.65 (0.88)	− 2.78**	0.12
4	0.46 (0.75)	0.42 (0.70)	0.71	0.05	0.55 (0.79)	0.57 (0.82)	− 0.19	0.02	0.44 (0.73)	0.56 (0.82)	− 5.02***	0.22
5	0.45 (0.70)	0.35 (0.64)	1.62	0.15	0.54 (0.72)	0.47 (0.75)	1.03	0.09	0.40 (0.68)	0.50 (0.82)	− 4.73***	0.21
Reactive	1.17 (1.50)	1.75 (1.99)	− 3.62***	0.32	1.21 (1.49)	1.90 (2.04)	− 4.51***	0.38	1.49 (1.82)	1.60 (1.86)	− 2.75**	0.12
6	0.65 (0.84)	0.95 (1.03)	− 3.53***	0.31	0.68 (0.83)	1.02 (1.05)	− 4.20***	0.35	0.82 (0.98)	0.88 (0.99)	− 3.24**	0.15
7	0.21 (0.54)	0.28 (0.68)	− 1.28	0.11	0.20 (0.52)	0.28 (0.63)	− 1.63	0.14	0.25 (0.62)	0.25 (0.60)	0.23	0.01
8	0.36 (0.61)	0.55 (0.84)	− 2.97**	0.25	0.36 (0.64)	0.61 (0.89)	− 3.85***	0.32	0.47 (0.75)	0.51 (0.81)	− 1.96	0.09

n = participants who during the last year have had regular contact with father (¹), mother (²), and with both father and mother (³); Psy = Psychological; Phy = physical; Fin = financial; C/D = control/domain.

*p < 0.05. **p < 0.01. ***p < 0.001.

Regarding the sex of the children, daughters exercise CPV toward their fathers and mothers for more reactive reasons than sons. Specifically, girls score significantly higher on the reasons related to their own temperament and in response to previous aggression from their fathers and mothers. In the case of sons, all instrumental reasons are more frequent in boys than in girls; however, the differences are not statistically significant in this sample.

As for the sex of the parents, increasing the arrival time at home, buying something the child wants, avoiding household or academic responsibilities, and the child’s own temperament are more frequent reasons for violence towards mothers than fathers.

Discussion

The purpose of this study was to provide a psychometric instrument to assess CPV in the Italian population and to know the frequency of this type of violence and the reasons for carrying it out in a sample of Italian adolescents.

The first objective was to examine the factor structure, reliability, and validity of the Italian version of the Child-to-Parent Violence Questionnaire (CPV-Q adolescents version;³⁶) in a sample of Italian adolescents. Hypothesis 1 is fulfilled by confirming in Italian adolescents the original CPV-Q structure as well as its excellent reliability and validity. This structure is made up of four factors (psychological, physical, financial, and control/domain violence) that are explained through 14 items using a parallel scale for the father and mother being one of the main contributions of the present study (see Supplementary Information File, “Appendix B”), which is also consistent with previous studies conducted with samples of Spanish³⁶, Chilean²⁷, Ecuadorian²¹ and Chinese⁴⁴ adolescents. Therefore, this shows that the CPV-Q is an instrument that allows assessing also in Italian adolescents CPV through the theoretical scrutiny of psychological, physical, financial, and control/domain violence^13,14,16,17.

As for the CPV-Q referred to mothers, the violence scale has an excellent fit, and the reasons subscale an acceptable fit. As for the CPV-Q referred to fathers, the violence scale also has an excellent fit, although in the reasons subscale, not all the fit indexes were optimally met. This could be because the reasons why children engage in violent behavior toward the father may have a degree of discrepancy or not be fully elucidated between the structural analysis model and the one used to reproduce the model, so more evidence is needed to verify the optimal fit of this subscale.

The factor loadings showed adequate values, indicating a strong association between the items and their factors. In all cases, they were greater than 0.50, suggesting a good representation of the latent construct⁶⁴, as in previous CPV-Q validation studies^27,36. The best factor loadings were found in the psychological and physical CPV factors, followed by the financial CPV factor. The factor loadings of the reasons subscale are also satisfactory. Only item 5 (“We watch what I want on TV”) of the control-domain factor was found to have a poor factor loading. It is possible that item 5 may need to be updated to the new technological changes that young people are experiencing and the new forms of control/domain towards parents. In this sense, it is necessary to revise this CPV-Q item to reflect the current forms of control and domain behaviors of children over their parents. It was checked that their elimination did not substantially affect the reliability and validity of the scale. For example, in the case of the father’s CPV, the CR could go from 0.48 to 0.54 and the AVE from 0.37 to 0.45, and in the mother’s CPV, the CR went from 0.53 to 0.55 and the AVE from 0.40 to 0.51. Consequently, we decided to keep this item from the original version of the CPV-Q to serve as a reference indicator to give way to a revised version of the instrument since the rapid social, technological, and relational changes justify an update of this item.

Good overall reliability for CPV-Q of the father and mother is highlighted as well as for the reasons subscale. Regarding the indices for each of the CPV-Q factors, the reference parameters for the nature of the data in this study (ordinal Alpha and Omega) are excellent in the psychological factor and acceptable in the financial and control-domain factors. The ordinal alpha and omega index were not calculated for the physical factor due to the low variability of responses. These items presented a floor effect. This phenomenon may be because we are evaluating adolescents from the community population, therefore, the behaviors included in the items of this factor are not accepted or are highly reprehensible. The low mean scores on these items are consistent with other studies in the Spanish-speaking population^21,27,36. In another study conducted with Chinese adolescents, good reliability was found in all four factors of CPV-Q⁴⁴, however, this study reports only traditional alpha.

The Average Variance Extracted (AVE) was found to be adequate in all factors of CPV-Q, except in the financial violence towards the father factor⁵⁷. Specifically, the AVE of the psychological and physical CPV factors stands out notably. This quantitative indicator explains that in the case of these two factors, more than 60% of the variance of the items is explained by the factor. Following Moral de la Rubia⁵⁷, the control/domain CPV items explain more than 37% of the variance of the factor, which is acceptable for 4-item scales, and the financial violence items explain more than 44% of the variance of the CPV toward the mother, which is acceptable for 3-item scales, but not toward the father (AVE = 0.40). In Burgos-Benavides et al.²¹ the AVE of the financial CPV was deficient toward both mother and father (AVE = 0.31). Cultural aspects could influence the differences in the results. For example, the interpretation of financial CPV may vary significantly between European and Latin American cultures. Therefore, the items that explain this latent construct may not be capturing adequate variance in the Latin American and Italian populations. Since this indicator is not presented by current⁴⁴ and previous studies^27,36 that have validated the CPV-Q in the adolescent population, it is not possible to establish further comparisons of our results. However, the present study shows the need to revise the items of this factor.

The HTM2 was the measure of discriminant validity used in the present study since it allows us to verify that the latent constructs measure different concepts from each other⁵⁹. Our results confirm that the four CPV factors present good discriminant validity, which is consistent with Burgos-Benavides et al.²¹. These results suggest that the CPV-Q can measure four crucial components of the theoretical model of CPV: those traditionally assessed in other CPV instruments, i.e., psychological, physical, and financial violence, and another fundamental component of CPV that is only assessed in the CPV-Q^15,34, i.e., control/domain behaviors. The reasons subscale also presents adequate discriminant validity, suggesting that the CPV-Q can assess both instrumental and reactive reasons. These distinctive features could significantly enhance the value and utility of this instrument in the field of study.

The second objective was to explore the measurement invariance of the CPV and its reasons based on the participant’s sex. The results confirm hypothesis 2. Burgos-Benavides et al.²¹ found total sex invariance in all the scales of the CPV-Q, except in the reasons for the CPV toward the father. In the present study, total invariance of the CPV-Q between the sex of the children was found in all scales, including the CPV reasons toward the father. That is, the structure, factor loadings, intercepts, and residuals are equal between both groups (sons and daughters). These results provide even more reliability to the comparisons between boys and girls in the frequency of CPV and its reasons. Although invariance is a fundamental property for studies comparing the equivalence of scores on a scale with multigroups, it is not always possible to compare the equivalence of scores on the same scale⁶⁵. This is because invariance is not an easy property to achieve. Therefore, another important contribution of this study is that having found evidence of total invariance benefits future studies that aim to compare CPV patterns as a function of the sex of the children.

The third objective was to analyze the frequency of the different types of CPV and the sex differences. High percentages of CPV are found in Italian adolescents, with psychological violence and control and domain behaviors being the most frequent, followed by financial and physical violence, as in previous studies (e.g., ^21,24,27). However, the absence of studies that have explored the magnitude of CPV in Italian samples limits the comparison of our results with similar samples. This, together with the high rates of CPV found, highlights the need for further research in the country. By types of CPV, the repeated frequency of psychological CPV in Italian adolescents (17.9–31.9%) seems to be slightly more frequent than in adolescents from other countries (3.1–28.8%;^{4,19,21–24,26,27}), whereas the occasional frequency of this type of CPV would be less frequent in the Italian context (53.8–64.10%) than in other contexts (41.3–95.3%). Physical CPV (9.7–14.1% occasional, 2.4–5.9% repeated) does not seem to show large variations in this sample compared to the percentages found in other samples (2–25% occasional, 0.9–5.4% repeated;^{4,19,21–24,26,27}). The pattern of financial CPV and control-domain behaviors in Italian adolescents shows a possible differential pattern compared to previous studies. Specifically, in the present study, the frequency of both occasional (27.6–35%) and repeated (8.4–11.9%) financial CPV would be less frequent than in other studies (31.4–50.9% occasional, 9.3–17.5% repeated;^21,24,28). In contrast, occasional (78.9–89.1%) and repeated (50.4–69.9%) control-domain behaviors would be more frequent in Italian adolescents than in those from other countries (53–66.6% occasional, 19.4–36% repeated;^21,24,27). It may be that the psychosocial and family profile of Italian adolescents who exert CPV differs from that found in other contexts, so our results underline the need for cross-cultural research to compare whether variables associated with CPV in other countries are the same or differ in Italian samples. This is important to determine whether CPV prevention and intervention programs developed in other sociocultural contexts can be applied in Italy or, on the contrary, specific designs for this population are needed.

Hypothesis 3, which refers to differences according to the children’s sex, is partially fulfilled. Specifically, daughters were expected to exert more psychological violence towards both parents. Our data show that Italian girls exercise more psychological violence towards fathers, in line with previous research^22,24, but not towards mothers, a result that does not coincide with previous^22,23,28 nor with those using the same instrument and criteria^21,24. Italian girls, however, exert more repeated physical violence towards mothers than boys, results that, again, do not coincide with the reference studies^21,24. Control and domain behaviors in this sample are also more frequent in girls than in boys, although in this case towards both parents, results are in line with expectations^21,24.

Hypothesis 4, regarding the differences based on the parent’s sex, is fully confirmed by the finding that all forms of CPV are more commonly directed at mothers than fathers, as in previous studies^23,24,27,28. Our findings provide further evidence to the abundant literature that consistently highlights the mother as the main victim of CPV (review by Simmons et al.⁸), which may further indicate that this pattern could persist regardless of the cultural context and the assessment tool employed. This underscores the importance of considering the influence of unequal gender socialization on the dynamics of CPV. It has been suggested that children may view mothers as less strong than fathers and may also direct more violence toward them simply because they are more accessible due to their traditional role as primary caregivers^66,67. However, current societal shifts toward gender equality and the increasing balance in parental roles between fathers and mothers could alter this dynamic. Likewise, it is also important to pay attention to the role of fathers, given the high levels of violence towards them found.

Regarding objective four related to the reasons for CPV, we expected more instrumental reasons in sons and more reactive reasons in daughters. Our hypothesis holds for girls but not for boys. In other countries, getting home later at night and avoiding academic responsibilities are more frequent reasons for exercising CPV in adolescent sons than in daughters (e.g., ^21,24). In the present study, these reasons in particular and the rest of the instrumental reasons in general are more reported by sons than by daughters; however, the differences are not significant, so further research is needed. In the case of girls, our study confirmed results from other countries in which reactive reasons are more frequent in girls than in boys^21,23,24,27, which could mean that girls from different cultures share a more reactive pattern when it comes to exercising CPV compared to boys. These same results have even been found in different profiles of CPV offenders³⁹. Difficulties in emotional regulation seem to be more common in girls who exercise CPV than in boys (e.g., ^22,39), so the lower emotional control could cause daughters to react with more anger to parental demands. Regarding the differences according to the sex of the parents, this study finds in a novel way that practically all the reasons analyzed are more frequent for violence towards mothers than towards fathers, in particular increasing the arrival time at home, buying something the child wants, avoiding household or academic responsibilities, and the child’s own temperament. These results could suggest that mothers may be more concerned with imposing rules and limits on their children than fathers, which would provide children with reasons for arguing with their mothers that could lead to CPV. Overall, the findings of this study show the need to analyze not only the reasons for CPV but also to do so in terms of the sex of aggressors and victims. This could be useful to better understand the family dynamics that influence adolescents’ motivations for CPV.

The study has some limitations to take into account for the interpretation of the results. The sample comes from two specific regions of Italy, limiting the generalization of the results to the Italian adolescent population. Future studies need to incorporate samples from different regions of the country and increase the sample size. The results are based on children’s self-reports, so to obtain a more complete understanding of the phenomenon, it is important to analyze the parents’ perspective. The results of the factor loadings of the control/domain CPV and the convergent validity of the financial CPV toward the father highlight the need to work on updating the items of these factors for a revised version of the CPV-Q. Despite these limitations, our findings are novel in the research field and provide unique empirical evidence for this population with important implications for research and professional practice.

Conclusion

This work provides the first valid and reliable instrument to assess CPV in Italian adolescents (CPV-Q). It is considered that the results of CPV-Q can provide valuable information leading to the selection of prevention and protection measures, both for the children and their parents. Our data suggest high rates of CPV in this population, justifying the urgent need to target resources to investigate this as yet unexplored phenomenon in the country. In this regard, professionals can use the CPV-Q to assess a wide variety of CPV behaviors and the reasons for the aggressions. It is necessary to identify the types of CPV present in each case but also the specific behaviors and the reasons associated with them, as the most appropriate therapeutic approach will depend on these questions. For example, it is important to focus on anger control in cases of reactive CPV, whereas if it responds to instrumental motivations, it is important to focus on the modification of dysfunctional beliefs and attitudes that justify the use of violence²⁴. The CPV-Q could also be used for preventive purposes as it is a brief and simple tool to detect cases of repeated CPV, potentially enabling early interventions. Another strategy to prevent CPV could be to provide adolescents, especially girls, with greater emotional regulation skills, as well as to break the possible cycle of bidirectional family violence, risk factors particularly related to reactive CPV⁶⁸. In any case, future studies should study the psychosocial and family profiles of CPV aggressors in this population. To our knowledge, this aspect has only been studied in Abazia’s study⁴³, which finds that authoritarian and permissive parenting styles are associated with CPV, and in Sicurella’s research⁴¹, conducted with 27 Italian professionals. The results show that exposure to violence is considered a constant risk factor, with many professionals interpreting adolescent violence as a symptom of deeper distress. Adolescents are seen as “wanting to be seen, not just watched,” often caught between intrusive, unaffectionate mothers and absent fathers who delegate caregiving responsibilities. Ineffective communication, role symmetry between parents and children, and families lacking clear boundaries are common characteristics. Children expect immediate gratification, are intolerant of frustration, and often respond aggressively when faced with restrictions⁴¹. It would also be necessary to make this phenomenon more visible and create greater social awareness in the country since the high frequency found could indicate a certain normalization in the emission of these types of behaviors⁶⁹. CPV is often downplayed, in part due to parents’ reluctance to report their children, making the issue socially imperceptible, as it remains hidden within the home. Reports of “abuse against family members or cohabitants” only represent the tip of the iceberg. Parents’ hesitance to report, driven by feelings of inadequacy and the desire to protect their children from stigmatization and criminalization, keeps much of the phenomenon concealed. To get a complete picture of the performance of CPV-Q and its usefulness for assessing CPV in different crops, it is suggested to evaluate this property so as not to limit the interpretation of results in different contexts. Also, to be able to compare the results of actual investigations in different cultures, it is suggested that additional reliability and validity indices of the CPV-Q should be provided. Finally, the CPV-Q should be validated in Italian young people and parents for a comprehensive understanding of the pattern of CPV in this population.

Author contributions

All authors have approved the version of the manuscript submitted. M.J.N-M: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Validation, Writing—original draft, Writing—review and editing; S.S: Conceptualization, Investigation, Supervision, Resources, Writing—original draft; R.S: Conceptualization, Data curation, Investigation, Methodology, Software, Supervision, Validation, Writing—original draft; L.B-B: Data curation, Formal analysis, Methodology, Software, Validation, Writing—original draft; M.C.C-L: Conceptualization, Data curation, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing—original draft, Writing—review and editing.

Funding

The authors declare that financial support was received for the publication of this article. This study was supported by the Department of Psychology of the University of Jaen and the Department of Sociology and Business Law of the University of Bologna under the chapters of fundamental research.

Data availability

The datasets generated during the current study are not publicly available due to the research is in the process of exploitation and dissemination of results by the team leading it but are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-93652-8.