The Relationship Between Kindness and Transgressive Behaviors in Adolescence: The Moderating Role of Self-Importance of Moral Identity

Abstract

Adolescence is marked by identity formation and moral development, often accompanied by increased transgressive behaviors. While existing research highlights the interplay between moral constructs and transgression in adolescence, the role of kindness remains underexamined. This study conceptualizes kindness as a multidimensional moral construct and investigates the relationship between different stages of kindness (i.e., egocentric, social/normative, extrinsically motivated, authentic) and transgressive behaviors among adolescents, also considering the moderating role of self-importance of moral identity. The participants were 215 Italian adolescents (aged 15–19) who completed a self-report questionnaire. The results showed that egocentric and authentic kindness were positively and negatively associated with transgression, respectively. Moreover, moral identity significantly enhanced the protective role of authentic kindness. These findings suggest that the relationship between kindness and transgression varies based on the stage of kindness and the importance adolescents attribute to their moral identity. They contribute to extending the understanding of kindness during adolescence, offering implications for reducing transgressive behaviors through targeted and innovative interventions.

1. Introduction

Adolescence is a critical developmental period during which profound physical, psychological, and social changes occur [1]. These changes enable adolescents to acquire new skills, such as thinking more abstractly, considering multiple points of view, and developing more organized and consistent self-descriptions, with personal and social values emerging as central [2]. Identity formation is one of the key developmental challenges of adolescence, often marked by exploration, doubt, confusion, moral dilemmas, and the pursuit of individuation and autonomy [3]—sometimes accompanied by transgressions of family and social norms. Theoretical conceptualizations [4] and empirical research [5] suggest that identity formation is somewhat related to transgressive or even antisocial behaviors throughout adolescence. During this period, the likelihood of engaging in transgressive behaviors such as rule-breaking, lying, or drinking alcohol significantly increases [6,7,8]. These behaviors may even become habitual because adolescents often see them as desirable, representing a way to access “adult life” and move beyond the status of childhood [9]. As Moffitt [4] stated, many adolescents perceive deviant behavior as a sign of strength and maturity, as well as a means of gaining acceptance from peers in both formal groups, such as classmates or teammates, and informal groups.

However, searching for adult status and peer acceptance may not fully explain why adolescents act transgressively. Although all adolescents have a strong desire to fulfill the basic need for relatedness [10] and to achieve a new social status [11], not all of them exhibit rebellious behaviors, and even fewer engage in antisocial ones. These behaviors result from multiple factors, which can act as either protective or risk factors [12]. Notably, they are closely associated with morality [13,14] to the extent that this association “has become an established finding” in the psychosocial research on transgression and delinquency [15,16], p. 290. As several previous studies have pointed out, adolescents who are more likely to engage in transgressive behaviors tend to exhibit a weaker moral identity [13], less developed moral judgment [15,17,18,19], and higher levels of moral disengagement—that is, a process that enables individuals to carry out actions that conflict with their moral principles [20]. In terms of Kohlberg’s stages of moral development, antisocial adolescents typically operate at stages 1 or 2 of moral cognition (i.e., pre-conventional morality, which focuses primarily on self-interest; [21]). Consistently, Nucci [22] found that aggressive adolescents (e.g., bullies) base their moral judgments on the anticipated reactions of adult authorities (e.g., teachers) and on their desire to avoid punishment, rather than recognizing the intrinsic value of moral principles and prosociality. In contrast, adolescents who do not engage in antisocial behaviors may attain stages 3 or 4 of moral development (i.e., conventional morality; [21]), where individuals are able to consider the impact of their actions on others [14,15].

Despite extensive evidence linking adolescents’ morality to transgressive or antisocial behaviors, the process by which this connection occurs remains unclear [14,23,24,25] mainly because of two reasons. First, the relationship between morality and deviant behaviors has been primarily investigated as linear and direct, often underestimating the role of possible intervening variables. Regarding this, recent evidence suggests a more complex pattern through which moral identity can shape prosocial or transgressive behaviors. Indeed, Lefebvre [26] highlighted that internal identity motivation is a robust cross-context predictor of engaging in moral acts and abstaining from immoral ones, while the predictive power of external identity motivation depends on the nature and context of the behavior. Moreover, Li et al. [27] pointed out that the relationship between moral sensitivity and prosocial behavior was mediated by moral disengagement and reciprocity norms. Finally, Wang et al. [28], involving a sample of adolescents, showed that moral identity—conceptualized as the extent to which being a moral person is central to one’s self-concept—was positively associated with prosociality. This relationship was mediated by feelings of belonging and experiences of moral elevation, a positive emotion that arises after witnessing virtuous actions and inspires a desire to act ethically [29]. Second, morality is an extensive concept encompassing morally relevant behaviors and thoughts [30], with no single agreed-upon definition in the psychological literature [31]. Several constructs, including moral identity, moral judgment, moral disengagement, virtues, and others, fall within the realm of morality. Many of them (especially the most cognitive ones) have been linked to adolescents’ deviant behaviors, though not always with convergent results [23,32,33,34]. However, researchers have investigated others less, or, to our knowledge, have not connected them at all to adolescent transgression.

Building on both the moral–cognitive developmental approach [21] and the social identity approach [35,36], this study aims to explore the relationship between adolescents’ kindness—a largely underexplored moral construct—and transgressive behaviors while also considering the potential moderating role of moral identity. There is scientific evidence to support studying moral identity as a moderator [37]. For example, in their study with adolescents aged 15–18 recruited across the U.S., Hardy et al. [13] found that the relationship between moral disengagement and aggressive behaviors was moderated by moral identity. Similarly, in a sample of Chinese adolescents aged 11–19, Yang et al. [38] highlighted that moral identity can buffer the negative impact of moral disengagement on cyberbullying behavior. Fan et al. [39] further showed that moral elevation reduces moral licensing (i.e., a phenomenon whereby engaging in initial moral conduct can pave the way for subsequent immoral behavior; [40]), particularly among individuals with low moral identity.

1.1. Kindness: A Multidimensional Moral Construct

During adolescence, kindness can play a pivotal role in self-definition and the formation of moral identity, as this is a period of life when individuals gradually strengthen their moral reasoning and engage with others’ perspectives and needs. Although it is easy to grasp the general meaning of kindness, finding its unequivocal definition in the scientific literature is a challenging task. It indicates a variety of attitudes, motivations, and behaviors involving care for others or a range of human qualities such as being helpful, altruistic, and polite. Several scholars across various scientific disciplines, such as psychology, sociology, and pedagogy, have defined kindness as a moral virtue or value [41]. According to Caldwell [42], kindness is inherently a moral duty to act for someone’s benefit that extends beyond legal responsibility. As such, being kind and showing acts of kindness would be based on morally grounded cognitive and emotional evaluations [43] and reflect emotions and internal states expressing benevolence [44,45,46].

Aligned with Caldwell’s definition [42], Comunian [47] analyzed kindness within a theoretical framework grounded in both moral and cognitive–developmental approaches [21]. According to this framework, kindness is not seen as a simple continuum from ‘entirely unkind’ to ‘extremely kind’, but rather as composed of four hierarchical developmental stages. In the first one, kindness is egocentric and subjective: individuals focus solely on their interests and benefits, disregarding others’ perspectives. During this period, acts of kindness are primarily performed to avoid trouble, satisfy personal needs, or achieve specific goals. In the second stage, kindness is defined as social/normative: individuals begin to consider others’ perspectives, but their actions remain primarily driven by their own interests and goals. In the third stage, instead, kindness is extrinsically motivated, largely influenced by external factors, such as the desire for social approval and acceptance. At this point, individuals take others’ perspectives into account and recognize the value of reciprocity and social harmony. In the final developmental stage, kindness is defined as authentic: individuals consider not only the perspectives of others but also broader societal and symbolic values.

In general, empirical evidence has highlighted the positive impact of kindness in various life domains. For example, engaging in kind behavior increases happiness and wellbeing [48]; it also reduces anxiety [49,50] and negative emotions [51]. Several recent studies report that kindness has a positive impact on prosociality, improving empathy, perspective-taking, compassion, and altruism [52,53]. In the context of group dynamics, the kindest individuals are more likely to gain popularity and achieve the highest status within the group [54]. However, the moral–cognitive developmental approach to the study of kindness suggests the possibility of variable relationships between kindness and behavioral outcomes, depending on the hierarchical stage an individual has reached [55]. It is likely that not all stages of kindness are associated with positive outcomes or function as protective factors against undesirable judgments, attitudes, or behaviors. Evidence of this comes from the only study to date that has examined the stages of kindness. Zagrean et al. [56] found that egocentric kindness may function as a risk factor for anti-immigrant prejudice among adolescents, whereas the most authentic form of kindness appears to serve as a protective factor.

1.2. Moral Identity: The Foundation of Moral Self-Conception

Moral identity is a relevant construct that lies at the intersection of moral development and identity formation [57]. Many aspects of social dynamics that might be involved in the formation of self-definition are morally relevant and based on different moral principles (e.g., “It is important to be kind because…”; [58]). Moral identity can generally be defined as the degree to which moral concerns (e.g., loyalty, justice, caring, etc.) are a central part of one’s identity and guide decision-making and actions across situations. Aquino and Reed [59] specifically referred to the “self-importance of moral identity” to describe the extent to which being moral is central to an individual’s self-concept. This concept encompasses the internalization of moral traits (i.e., the personal relevance of possessing them) and the symbolization of these traits to others (i.e., the outward display of moral traits and behaviors). Indeed, moral identity encompasses both a private dimension of the moral self, which reflects the subjective experience of morality, and a public dimension, through which the display of moral behaviors serves as a means of affirming one’s morality.

Moral identity can be conceptualized as a significant source of moral motivation [60] and is associated with a broad range of values, beliefs, attitudes, and behaviors, particularly when it holds high self-importance [61,62]. A meta-analysis conducted by Hertz and Krettenauer [63], involving samples from adolescence to adulthood, showed a significant and positive relationship between moral identity and prosocial behaviors, as well as the avoidance of antisocial and unethical behaviors. Krettenauer et al. [64] found that the self-importance of three moral attributes—honesty, fairness, and caring—was positively correlated with feelings of integrity and compassion. Similarly, Hardy et al. [13] found that adolescents who placed greater importance on their moral identity exhibited higher levels of prosocial behaviors (e.g., civic engagement and charity) and lower levels of transgressive behaviors (e.g., aggression and rule-breaking). There is also evidence that individuals for whom moral identity plays a central role in self-definition tend to evaluate others’ helping behaviors as more caring, moral, and socially responsible [61]. In their research examining whether individuals with a highly self-defining moral identity are more susceptible to experiencing moral elevation after being exposed to acts of exceptional moral goodness, Aquino et al. [65] reported a strong link between moral identity, witnessing acts of goodness, and engaging in prosocial behaviors. In line with these findings, recent studies have shown that moral identity is closely linked to moral elevation. Specifically, individuals with strong moral values are more likely to experience positive feelings such as moral elevation, which, in turn, fosters prosocial behavior [28,66,67].

Furthermore, the importance assigned to moral identity is a significant moderator in many relationships, generally acting as a buffer against the impact of risk factors (e.g., Machiavellian leadership, job insecurity, sport supplement use) on undesirable behaviors (e.g., opportunistic behaviors, unethical pro-organizational behaviors, doping; [68,69,70]). There are also studies showing that a strong moral identity can enhance the positive relationship between desirable variables, such as ethical attitudes and ethical conduct [71].

1.3. The Present Study

Based on all the above findings and considerations, the main aim of this study was to analyze the association between the four stages of kindness proposed by Comunian [47] and transgressive behaviors in adolescence. We expected that the four stages of kindness would have different relationships with transgression. Drawing on previous findings [56], we hypothesized that egocentric kindness would be positively associated with transgressive behaviors (H1). In contrast, we hypothesized that authentic kindness would be negatively associated with transgressive behavior (H2). To better understand the relationship between kindness and transgression, we considered the moderating role of adolescents’ self-importance of moral identity. Indeed, moral identity can be considered a motivational construct that determines the salience of moral values in guiding behavior [60]. Consistently, several previous studies have highlighted the buffering role of moral identity in shaping the relationship between moral dimensions and antisocial or prosocial outcomes [13,68,71]. Specifically, we hypothesized that moral identity would weaken the association between egocentric kindness and antisocial behaviors (H3) while enhancing the protective role of authentic kindness against such behaviors (H4).

Regarding the other two forms of kindness, described above as social/normative and extrinsically motivated, given their intermediate positioning in moral and cognitive development [47] and the lack of previous research results, we were unable to formulate any specific hypotheses, preferring to adopt a more exploratory approach.

2. Materials and Methods

2.1. Participants and Procedure

The participants were 215 adolescents (77% girls) aged between 15 and 19 (M = 16.7, SD = 0.72). We determined the sample size based on an a priori calculation using G*Power 3.1 software. The analysis indicated that a minimum of 178 participants was required to detect a medium effect size (f² = 0.15) in a linear multiple regression with eleven predictors, assuming an alpha level of 0.05 and a statistical power of 0.95.

Ninety-eight percent of participants were born and raised in Italy and lived in Italy at the time of data collection. In total, 43.3% and 48.8% of them attended the third and fourth grades of high school, respectively. A total of 76.3% of adolescents lived with both parents, while 18% lived only with their mother, 1.9% with their father, and 3.8% selected “other” as their living arrangement.

The participants were recruited in collaboration with 19 upper secondary schools across Italy. Of these, ten were high schools (52.6%), including six scientific high schools; two classical high schools, and two humanities-oriented high schools. The remaining nine schools (47.4%) were technical or vocational institutes. Most schools (80.8%) were in central Italy and were attended by students from diverse educational and socioeconomic backgrounds. Within each school, classes were randomly selected for participation. Students in the selected classes were invited to attend a preliminary online meeting, during which the principal investigator explained this study’s objectives, procedures, and participants’ rights. Although we did not systematically record the distribution across grade levels, the sample included students from both the early and later years of upper secondary education. After being informed about the voluntary and free nature of participation, the students—or their parents in the case of minors—provided written informed consent. Those who consented and whose parents authorized participation completed an anonymous online questionnaire in their classrooms during school hours in the presence of a teacher and a research staff member.

This study was approved by the Ethics Committee of the LUMSA University (protocol No. 7/2021) and it was conducted following the Declaration of Helsinki and with the ethical guidelines for research provided by the Italian Psychological Association (AIP). The first author of this manuscript has previously completed the National Institute for Health training course “Protecting Human Research Participants” (Certification Number: 2868994).

2.2. Measures

Other than questions on key sociodemographic data (i.e., sex, age, country of birth, nationality, cohabiting family members, and year of enrollment in high school, technical, or vocational institutes), the questionnaire administered to the adolescents included the Kindness Scale [47], the Self-Importance of Moral Identity scale [59], and the Self Report of Antisocial Behavior [6,9]. This study partially reused data previously published in Zagrean et al. [56]. Specifically, the variables gender and the four stages of kindness were retained from the earlier dataset. All other variables included in the current study, namely, transgressive behaviors [6,9] and the self-importance of moral identity [59], were not part of the previous publication. Furthermore, the research questions, hypotheses, and statistical models are entirely original to the present work. The questionnaire also included some indications and questions to reduce the risk of careless responding bias. Specifically, following the guidelines provided by Ward and Meade [72], we reported the effort and time required to complete the questionnaire and asked the participants to read and agree to three statements related to their commitment to the survey (i.e., “I acknowledge that this study will take approximately 20/25 min”; “I am aware that my participation will contribute to the advancement of scientific knowledge”; “I undertake to read each question carefully and answer truthfully”) [72]. Additionally, we included a control item in the middle of the survey (i.e., “This is a control item. Please select absolutely agree”).

2.2.1. Kindness Scale [47]

The scale was originally developed in Italian and has shown good reliability and construct validity. It consists of 20 items divided into 4 subscales, each corresponding to a different type of kindness, based on Kohlberg’s stages of moral development (Kohlberg, 1984) [21]. To provide a detailed evaluation of the internal consistency, we reported both Cronbach’s alpha and McDonald’s omega for each subscale.

-

Stage 1 (egocentric): kindness is driven solely by personal interest and is considered selfish and subjective (“I am a kind person because it is easier to get what I want”, α = 0.89; Ω = 0.88).

-

Stage 2 (social/normative): kindness is viewed as recognizing reciprocal relationships, though it is still motivated by self-interest (“I am kind when it serves any purpose”, α = 0.58; Ω = 0.60).

-

Stage 3 (extrinsically motivated): kindness is extrinsically motivated, as people begin to consider the perspectives of themselves and others (“I am kind so that I seem as nice as others”, α = 0.65; Ω = 0.66).

-

Stage 4 (authentic): kindness is based on mutual respect and societal perspective, where individuals believe in respecting others (“I am kind because I believe in respecting the dignity of others”, α = 0.77; Ω = 0.78).

The participants rated each statement on a 4-point Likert scale (1 = not at all true to 4 = completely true). The score for each stage of kindness was calculated by averaging the responses to the respective items, with higher scores indicating a higher level of kindness. All original items were retained, as item removal did not result in any meaningful improvement in internal consistency. While the internal consistency for Stage 2 and Stage 3 was relatively modest (α = 0.58 and α = 0.65, respectively), these values fall within an acceptable range for scales assessing complex and multifaceted psychological constructs [73].

2.2.2. Self-Importance of Moral Identity [59]

This scale measures the importance that respondents assign to a series of moral traits, and, for this study, the items were back-translated into Italian. The participants were presented with a list of nine moral characteristics (e.g., caring, compassionate, fair) and were asked to think about a person who holds those traits. Then, they rated the extent to which they would like to be someone who holds and expresses those characteristics by responding to 10 items on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree). Examples of items include the following: “Being someone who has these characteristics is an important part of who I am” and “I am actively involved in activities that communicate to others that I have these characteristics.” The total score was calculated by averaging the items: the higher the score, the higher the importance of moral identity (α = 0.72; Ω = 0.75).

2.2.3. Self-Report of Antisocial Behavior [6,9]

This scale, in its original Italian version, was used to measure adolescents’ propensity for transgressive behavior. The respondents were asked to indicate how often in the previous week they engaged in rule-breaking behaviors, such as lying to parents, hitting someone, stealing something from a store, doing graffiti on public transportation or property, and using alcohol. The scale consists of 18 items, rated on a 4-point Likert scale (1 = never to 4 = often). Examples of items include the following: “Have you stolen something from a store?” and “Did you swear at anyone?”. The scale is unidimensional, and the total score was calculated by averaging the items: the higher the score, the higher the frequency of transgressive behavior (α = 0.86; Ω = 0.88).

2.3. Data Analysis

Preliminarily, we provided a descriptive analysis of this study’s variables, reporting means (M), standard deviations (SDs), ranges, skewness, kurtosis, and linear correlations (Pearson’s coefficients).

Then, we tested a moderated regression model to explore the relationships between the four stages of kindness and transgressive behavior and to assess whether and how the self-importance of moral identity moderated these associations. We included age and gender as control variables. Moderation implies that the existing relation between two variables (in this case, kindness and transgressive behavior) changes as a function of a third variable (in this case, moral identity; [74]). The continuous predictors were grand-mean-centered to simplify the coefficients’ interpretation and try to reduce multicollinearity [75]. We considered a p-value ≤ 0.01 to be statistically significant for this analysis to limit the likelihood of Type I errors. Further interpretation of the significant interactions was achieved by analyzing the regression slopes [76].

We conducted the preliminary analysis and tested the regression model using the software SPSS Statistics version 23 (IBM Corp: Armonk, NY, USA. Released, 2015). Additionally, we used the software Interaction! [77] to analyze the slopes in the case of significant moderating effects.

3. Results

To test our hypotheses, we conducted a moderated regression analysis examining the relationships between the four stages of kindness, moral identity, and transgressive behaviors. We first present descriptive and correlational statistics, followed by the results of the moderated regression analysis.

3.1. Descriptive Statistics

Table 1. Descriptive statistics and correlation coefficients.

Variables	M	SD	Range	Skewness	Kurtosis	1.	2.	3.	4.	5.	6.
1. Egocentric kindness	1.73	0.76	1.00–4.00	1.00	0.47	−	0.28 **	0.10	−0.22 **	−0.15 *	0.38 **
2. Social/normative kindness	2.95	0.51	1.60–4.00	−0.19	−0.01		−	0.35 **	0.27 **	0.06	0.08
3. Extrinsically motivated kindness	2.94	0.53	1.60–4.00	0.29	−0,18			−	0.52 **	0.37 **	−0.10
4. Authentic kindness	3.27	0.48	2.00–4.00	−0.13	0.71				−	0.47 **	−0.29 **
5. Self-importance of moral identity	3.47	0.56	1.80–5.00	0.16	0.17					-	−0.16 **
6. Transgressive behaviors	1.82	0.49	1.00–4.00	0.49	1.66						-

Note. ** p < 0.01, * p < 0.05.

presents the descriptive and correlational results.

As shown in Table 1, extrinsically motivated and authentic kindness were significantly and positively associated with the self-importance of moral identity. Egocentric kindness was slightly negatively related to the self-importance of moral identity. Additionally, egocentric kindness was positively linked to transgressive behaviors, while authentic kindness showed a negative association. Lastly, the self-importance of moral identity was significantly and negatively associated with transgressive behaviors.

3.2. Moderated Regression

Before conducting the moderated regression analysis, we evaluated key assumptions to ensure the appropriateness of the model. We assessed the normality of residuals by inspecting the Q-Q plot and the histogram, both of which suggested that residuals were normally distributed. This was further supported by a non-significant Kolmogorov–Smirnov test (p = 0.379). As expected, some degree of multicollinearity was present (VIF values > 10; tolerance < 0.1), likely due to the conceptual and empirical proximity between kindness and moral identity, both of which are rooted in prosocial moral functioning. Although the variables are empirically correlated, their theoretical foundations remain distinct. The correlation coefficients, ranging from 0.40 to 0.48, reflect weak-to-moderate relationships according to conventional thresholds [78], supporting the decision to include them as separate predictors.

We also assessed the potential impact of common method bias due to the self-report nature of the data by conducting Harman’s single-factor test. The first unrotated factor accounted for only 17.8% of the total explained variance (65.13%), which is below the 50% threshold commonly used to indicate the presence of substantial common method variance [79]. This suggests that common method bias is unlikely to represent a serious concern in our data.

The moderated regression analysis presented in

Table 2. Moderated regression analysis results.

	Transgressive Behaviors
			95% CI
Predictors	B	SE	Lower	Upper	β	p
Age	0.02	0.05	−0.07	0.11	0.03	0.70
Gender	−0.06	0.08	−0.22	0.09	−0.06	0.41
Egocentric kindness	0.20	0.05	0.11	0.29	0.31	0.01
Social/normative kindness	0.04	0.07	−0.09	0.18	0.04	0.56
Extrinsically motivated kindness	−0.03	0.07	−0.18	0.11	−0.04	0.66
Authentic kindness	−0.24	0.08	−0.41	−0.07	−0.24	0.01
Self-importance of moral identity (SIMI)	0.03	0.07	−0.11	0.16	0.03	0.71
Egocentric kindness × SIMI	−0.09	0.08	−0.25	0.07	−0.19	0.27
Social/normative kindness × SIMI	0.09	0.13	−0.16	0.35	0.05	0.48
Extrinsically motivated kindness × SIMI	0.33	0.14	0.05	0.61	0.21	0.02
Authentic kindness × SIMI	−0.43	0.16	−0.74	−0.12	−0.24	0.01

showed that the model was significant, F (9205) = 6.334, p < 0.001, explaining 18.3% of the variance of transgressive behaviors. The inspection of the regression coefficients indicated that both egocentric kindness and authentic kindness were significantly associated with transgressive behaviors. Specifically, higher levels of egocentric kindness were positively associated with increased transgressive behaviors (β = 0.31, p = 0.01), whereas higher levels of authentic kindness were negatively related to such behaviors (β = −0.24, p = 0.01).

Furthermore, we observed a significant interaction between authentic kindness and the self-importance of moral identity (β = −0.24, p = 0.01), indicating a small-to-moderate effect size. Simple slope analysis (Figure 1) revealed that at high levels of self-importance of moral identity, the negative association between authentic kindness and transgressive behaviors was stronger (simple slope = −0.35, SE = 0.10, 95% CI [−0.55, −0.15], p < 0.01) compared to low levels of self-importance of moral identity (simple slope = −0.21, SE = 0.09, 95% CI [−0.40, −0.02], p < 0.05).

4. Discussion and Conclusions

This study examined the relationship between adolescents’ kindness, conceptualized using the hierarchical framework proposed by Comunian [47], and their engagement in rule-breaking behaviors, such as stealing, lying, swearing, and drinking alcohol. It also explored the moderating role of the self-importance of moral identity in shaping this relationship. Indeed, adolescence is a fundamental phase for identity formation and moral development, often accompanied by increased rebellious behaviors, since they might symbolically represent a way to enter “adult life”. The existing literature has documented the interplay between moral constructs and transgression in adolescence, but the role of kindness, especially in its multidimensional nature, needs to be further explored. Additionally, moral identity significantly reduces the impact of risk factors, mitigating the possibility of adopting undesirable behaviors [68,69,70]. In the following section, we interpret the findings of the present study in relation to the existing literature on the connection between morality and transgressive behaviors in adolescence.

Our findings show that different stages of kindness are distinctly associated with patterns of rule-breaking behavior. Consistent with our hypotheses (H1, H2), egocentric kindness was positively related to transgressive behaviors, while authentic kindness was negatively associated with them. Egocentric kindness, characterized by self-interested or self-serving motivations, may—whether intentionally or unintentionally—lead to behaviors that prioritize personal gains over adherence to social norms [56]. Adolescents displaying high levels of egocentric kindness often lack the perspective-taking abilities needed to appreciate the broader social consequences of their actions. In this context, transgressive behaviors may serve as a strategy for fulfilling developmental needs, such as gaining peer acceptance or asserting strength and maturity [4,9].

In contrast, authentic kindness, rooted in genuine concern for others, aligns with prosocial values that discourage harmful actions. Accordingly, Malti [46] argued that authentic kindness, characterized by advanced moral reasoning and altruistic motivations, plays a pivotal role in shaping moral behavior. The value of reciprocity, which we may believe to be in contrast with the adoption of antisocial behaviors, is recognized as Stage 4 of kindness, namely, the authentic one [47]. Consistent with our findings, this form of kindness has the potential to inhibit transgressive behaviors. This result aligns with findings by Zagrean et al. [56] that underscore that authentic kindness not only reduces prejudice but also promotes empathetic and inclusive behaviors. Thus, it can be speculated that adolescents who reach this stage of kindness may exhibit greater awareness of the role social norms play in fostering and sustaining positive and harmonious relationships. Furthermore, while the desire for social acceptance may drive earlier stages of kindness, authentic kindness is primarily guided by intrinsic values. This intrinsic motivation enhances adolescents’ resilience to social pressures that might otherwise lead to deviant behaviors.

The results for social/normative kindness (Stage 2) and extrinsically motivated kindness (Stage 3) were instead less conclusive, consistent with the exploratory nature of our approach to these kindness stages, which did not allow us to formulate specific hypotheses. Future research should further analyze them in relation to other moral variables (e.g., emotional dimensions of morality or cultural values), as they may be typical of transitional phases in moral development with varying susceptibility to personal or contextual influences. Future research should also try to comprehend whether and to what extent the relationships here investigated may vary depending on adolescents’ gender, as the existing literature suggests that boys and girls may express kindness or morality differently [3,13].

Interestingly, the results showed that the self-importance of moral identity significantly moderates the association between kindness and transgression, specifically strengthening the protective role of authentic kindness (as predicted in H4). In contrast, it did not mitigate the risk posed by egocentric kindness (contrary to H3). This seems to confirm the relevance of moral identity as a core component of self-conception, able to work as a self-regulatory process that motivates moral actions [39,57,59]. The theoretical framework by Aquino and Reed (2002) further emphasizes that moral identity serves as a regulatory mechanism, motivating alignment between moral values and actions [59]. In this regard, Nucci (2002) highlights the developmental progression of moral reasoning in adolescence [22]. Adolescents who progress to higher stages of moral development are better equipped to evaluate the consequences of their actions, leading to reduced engagement in norm-violating behaviors. Adolescents with a strong moral identity are usually characterized by a sense of agency and responsibility, leading them to desire and aspire to a high level of internal consistency between their moral standards and behaviors [80]. In this context, it seems plausible that moral identity activates the role of kindness, in its more authentic form, in shaping moral behavioral outcomes. However, moral identity does not buffer the risk posed by egocentric kindness, as behaviors at this stage are predominantly driven by extrinsic motivations [47]. One possible explanation for the absence of a moderating effect may refer to the nature of egocentric kindness itself, which is driven by external and self-focused motives and may not activate the internalized moral self. As suggested by Fan et al. [39], individuals with low moral identity are more susceptible to moral licensing effects, whereby initial prosocial acts, especially if extrinsically motivated, can paradoxically justify subsequent transgressive behaviors. Moral identity, which is intrinsically linked to a profound self-concept and moral principles, cannot find fertile ground for influencing behaviors rooted in personal gain. This also aligns with the work of Lefebvre [26], who has suggested that internal motivation is a strong cross-context predictor of engaging in moral and avoiding immoral behaviors, whereas external motivation’s predictive power is contingent on the nature and context of the behavior.

The present study sheds light on important associations involving kindness. Nevertheless, it has some limitations that must be acknowledged. First, the cross-sectional design prevented us from making causal inferences based on the results. Moreover, our correlational analysis highlighted significant associations between moral identity and authentic kindness, suggesting the potential for further hypotheses regarding mediation. However, given the cross-sectional nature of the data and the theoretical risk of reverse causality, conducting such an analysis would not provide reliable causal insights [81]. Future studies should address this limitation by adopting a longitudinal approach to trace the developmental trajectories behind kindness and moral identity. Secondly, due to the conceptual overlap among the four kindness dimensions and moral identity, multicollinearity among predictors was detected. Although the two constructs are theoretically distinct, and our main interest lies in the moderating role of moral identity, this issue suggests that some caution is warranted when interpreting the unique contribution of each predictor. Thirdly, as all variables in the present study were assessed through self-report measures, future research would benefit from integrating multi-method approaches (e.g., behavioral assessments) to reduce potential method bias and more comprehensively capture the complexity of adolescents’ moral developments. Finally, the convenience sampling procedure resulted in a higher proportion of girls compared to boys. Moreover, most of the participants were recruited from high schools in central Italy. Although we involved several high schools, ensuring the participation of students from diverse backgrounds, future studies should involve high schools throughout the national territory to strengthen the generalizability of the results. Furthermore, exploring the role of contextual factors, such as family environment and influences, could provide a more nuanced understanding of the relationships between kindness, moral identity, and behavioral outcomes.

Despite these limitations, to the best of our knowledge this is the first study to explore kindness as a protective factor against transgression during adolescence, offering novel insights into its role in connection to adolescent moral development. It provides several theoretical and practical implications. First, it contributes to the development of the literature on adolescents’ morality by incorporating a multidimensional view of kindness and emphasizing the variability in its relationship with transgression depending on the stage of moral development. The most intriguing finding is, indeed, that the various facets of kindness contribute differently to shaping transgression—and interact differently with the self-importance of moral identity—highlighting the complexity of kindness as a multidimensional construct in moral development. Additionally, the moderating role of moral identity highlights the importance of targeting identity formation in interventions, fostering a strong alignment between adolescents’ moral virtues and actions. Perkins et al. [53] provided evidence of the psychological and social benefits of kindness-based interventions. Their work supports our hypothesis that authentic kindness can significantly enhance adolescents’ resilience against negative peer influences and can promote prosocial behavior. More specifically, we suggest that educational and community interventions aimed at reducing transgressive behaviors among adolescents could benefit from strategies that promote the intrinsic value of kindness and enhance the salience and centrality of moral identity in self-concepts. Several prevention interventions focused on kindness and positive behavior among adolescents have shown promising results [82]. Nevertheless, none of them combine kindness, differentiating it in its various stages with moral identity. Kindness, particularly when authentic, serves as a resource, while moral identity can act as a critical moderator, shaping the expression of kindness in behaviors and yielding lasting behavioral benefits. Programs that integrate kindness-based strategies with moral identity development could offer long-lasting behavioral benefits, promoting greater consistency between adolescents’ moral virtues and actions. The question then becomes: how can we intervene to ensure that kindness is authentic and moral identity is salient? Future interventions should be designed to not only foster prosocial behavior, but also promote adolescents’ self-reflection on their values and identity formation. We suggest that practical strategies aimed at fostering authentic kindness and its integration within adolescents’ moral identity could include structured group reflections and guided discussions. These activities might encourage adolescents to reflect on how they feel and what they take away from situations in which they act with genuine kindness, even when such actions are not rewarded or socially visible. The goal here is to support adolescents in recognizing authentic kindness not simply as a behavior to be performed, but as an expression of who they are or aspire to be. These strategies are consistent with the core principles of Social and Emotional Learning (SEL), which highlight the importance of self-awareness, empathy, and responsibility. Incorporating these elements into classroom practice not only promotes adolescents’ moral development but also supports broader socio-emotional growth.

In conclusion, this study underscores the complexity and significance of kindness as a multidimensional construct in adolescent moral development. By targeting both authentic kindness and moral identity, future interventions could empower adolescents to navigate social challenges with resilience, fostering a generation more committed to collective coexistence and well-being.