Comparative study of the usage of Nonviolent Communication (NVC) and Restructuring Cognitive Distortion (RCD) Education program for understanding and dealing with problem-solving, Emotional Intelligence, and Resilience

Abstract

This study investigated the effects of two interventions—Nonviolent Communication (NVC) and Restructuring Cognitive Distortion (RCD)—on adolescents’ problem-solving, emotional intelligence, and resilience. The study followed a quasi-experimental pretest–posttest design. A total of 72 adolescents were initially allocated (24 per group) using stratified convenience procedures. Participants were equally allocated into two experimental groups (NVC and RCD) and one control group, ensuring demographic balance across the groups. Post-test completers formed the final analytic sample (N = 49; NVC = 18, RCD = 16, Control = 15). Analyses therefore used unequal group sizes. After participating in a pre-test, two experimental groups were structured by eight weekly sessions of NVC and RCD program, for 90 min each session. They were all participated in a post-test to examine the probable effects. Problem-solving, emotional intelligence, and resilience were measured using the Problem-Solving Inventory (PSI), the Trait Emotional Intelligence Questionnaire–Adolescent Short Form (TEIQue-ASF), and the Connor–Davidson Resilience Scale (CD-RISC). Results indicated that adjusting for pre-test scores, ANCOVAs (Holm–Bonferroni–corrected) showed significant group effects for the Problem-Solving Inventory (PSI) total and the Personal Control subscale: both intervention groups outperformed the control group, with the NVC arm exhibiting the highest adjusted post-test means; differences between NVC and RCD did not reach statistical significance. No between-group differences emerged for the other PSI subscales. For trait emotional intelligence (TEIQue-ASF), between-group effects were not significant. For resilience (CD-RISC), no significant group effects remained after correction. These findings suggest that participation in NVC was associated with greater improvement in problem-solving outcomes relative to control, while between-group effects on emotional intelligence and resilience were not significant. Communication-focused training (NVC) appears advantageous for problem-solving outcomes relative to no intervention; evidence for advantages over RCD and for effects on emotional intelligence or resilience is inconclusive and warrants further study.

Introduction

In psychological terms, the human ability to sustain well-being, mental health and adaptive functioning over time despite or due to internal (e.g. genetic vulnerabilities) and external challenges (e.g. environmental or social stressors) is described in terms of sustainability. This also focuses not merely on surviving, but thriving, or flourishing, in adversity. The paper illustrates how individual psychological resilience strategies are developed through extended interaction with the environment, including social and psychological environments. In this sense, psychological sustainability is broader in focus than classical notions of sustainability, which tends to focus on just survival, with little scope for growth or recovery purloined or otherwise lost. These observations have important consequences from a psychological perspective, and for devising plans of action aimed at enhancing mental health [1, 2].

Adolescence is a critical stage of emotional development, marked by heightened emotional intensity and frequent fluctuations (Steinberg, 2020). Effective regulation of emotions during this period is essential for resilience, as adolescents who can balance their feelings are better prepared to cope with stress and adversity in adaptive ways [3]. Moreover, the ability to generate constructive solutions to problems is associated with greater emotional well-being and social competence [4]. Supportive social environments, including parents and peers, play a crucial role in fostering emotional regulation strategies and coping mechanisms, which in turn promote lifelong psychological health and adaptive functioning [5].

Discussing adolescent well-being naturally presumes the concept of emotional intelligence (EI). Salovey and Mayer [6] have proposed that EI consists of four fundamental ability components: perception of emotions, using emotions to facilitate thought, understanding emotions and managing emotions. Recent studies have shown the importance and complexity of these abilities in relation to mental well-being [7]. For instance, Fiori et al. [8, 9] introduced Emotion Information Processing (EIP)—the ability to respond to and correlate with emotional information—distinct from Emotional Intelligence Knowledge (EIK), or knowledge and understanding of emotions and their functions. This difference provides an insight into how emotions are processed as well as how they influence behavior and decision making. Furthermore, Pérez-Fernández et al. [7] highlighted the role of EI as a protective factor against the development of emotional disorders and in enabling behavior regulation, hence reducing the risk of depression and anxiety.'' In addition to EI, an effective communication technique, NVC as proposed by Rosenberg [10] also seems to be a powerful communicative vehicle for the expression of emotions and needs without judging or blaming others, which helps to express the emotion and requests. It leads to compassion and clarity in communication. Wherein EI means recognizing, employing, comprehending, and handling emotions,however, NVC translates those emotional skills into effective speaking and more positive connections.

Moreover, cognitive distortion can impede the proper application of emotional intelligence. The result is distorted emotional response and reduced emotional regulation due to maladaptive thoughts such as catastrophizing or over-generalization. A study by Goh et al. [11] demonstrated that participants with high levels of cognitive distortions (CDs) were likely to misperceive feelings or act impulsively as an obstacle to EI. This finding suggests that cognitive distortions may compromise the awareness of emotions, which may interfere with emotions and relational patterns. Also, research has found that individuals engaging in cognitive distortions at a high level have problems with empathizing with others and accurately reading others’ emotional cues [12].

Research conducted by psychological scientists suggests that emotional intelligence (EI) significantly influences sustainable development. When individuals participate in educational programs designed to enhance EI, they not only strengthen their emotional growth and interpersonal relationships but also develop greater social awareness, empathy, and responsibility. These improvements can foster cooperative behaviors, empathetic decision-making, and long-term thinking—traits essential for responding effectively to environmental challenges. Empirical research supports this link: for example, a systematic review showed associations between dimensions of EI and adolescent pro-environmental behavior [13].

Another study demonstrated that trait EI enhances emotional connectedness to nature, which in turn mediates engagement in pro-environmental behaviors [14]. Similarly, a study among university employees found that EI mediates the relationship between environmental intentions and actual sustainable behaviors [15, 16]. Thus, EI contributes indirectly to sustainable development by shaping individuals who are more empathetic, socially responsible, and behaviorally inclined toward environmental stewardship.

Problem-solving is a core competency that supports adolescents’ well-being and it should be deliberately cultivated throughout education. Recent work in undergraduate biology shows that students’ approaches to problem-solving are organized around five interacting aspects—knowledge, strategy, intention, metacognition, and mindset—which together yield qualitatively different approaches to solving problems [17]. In particular, metacognitive regulation (planning, monitoring, evaluating) and students’ self-efficacy processes such as “self-coaching” help learners persist through challenging tasks [18]. Moreover, explicit instruction that targets metacognition enhances learning and problem-solving in biology courses, underscoring the value of embedding these practices in curricula [19].

One of the interesting psychological tools that affect problem-solving is Nonviolent communication (NVC). In a study by Espiritu, [20], it was proved that NVC promotes empathetic conversation, focusing on the needs of either the speaker or the listener to solve problems effectively. Another study showed that the emphasis of NVC is neither on blaming nor on fault-finding, but it focuses on mutual constructive problem-solving to provide a tool for resolving conflicts peacefully [21].

Cognitive distortion is also a significant psychological variable associated with problem-solving. Such cognitive distortions—like over-generalization and catastrophizing—may limit the ability to think rationally, resulting in maladaptive decision making [22]. Moreover, it was stated that the higher cognitive distortions are related to a tendency to use maladaptive problem-solving strategies to avoid jumping to conclusions [23].

High level of thinking skills may also contribute to sustainability. People who are good at problem-solving have the ability to produce innovative and impactful solutions for personal problems (and, I would venture, societal ones). Trainings such as Non-violent Communication (NVC) contribute to the development of the capacity among adolescents to know for what they stand, what their needs are, how they can address these needs with strategies that open dialogues [24]. By nurturing these capacities, NVC assists adolescents in improving positive problem-solving habits that contribute to personal development, as well as promoting action and decision making in line with objectives of sustainable development.

Recently studies focus more on resilience from its various dimensions to highlight the significance in mental health and general well-being. For instance, one study examined the relationship between mental resilience and health outcomes. It showed that increasing resilience has negative relationship with the risk of committing suicide (Mendes et al., 2024). In addition, resilience has been considered as a protective factor during times of adversity when regular resources are inadequate, helping people cope with following difficulties [25]. These results suggest that promoting resilience may help increase life satisfaction, and decrease mental health problems.

Regarding that, increasing the level of resilience can enable the adolescents to develop an appropriate perception of the failed situation and attempt to cope with that situation minimizing the communication gap. The findings of such studies reveal that NVC has a significant effect on the level of resilience. Therefore, the literature demonstrated that a strong level of resilience may play a crucial role in sustainable development [26].

Cognitive distortions can decrease resilience by promoting negative thought processes and influences of pessimism. Studies suggest that individuals with high cognitive distortions are less likely to report self-efficacy, impaired psychological resilience toward dealing with recovery capacity [27]. Maladaptive cognitive styles including dichotomous thinking and catastrophizing, can decrease the capacity to treat stressors or stressors’ responses as manageable, undermining the acquisition of resilience (Bruce & Smith, 2017). Specifically, it is true for adolescents whose cognitive and affective regulation capacities are still muted and they are at a risk for the impacts of an unbalanced thinking process.

This study aims to find out the effectiveness of two psychological education programs (NVC and RCD) on adolescents regarding the psychological concept of problem-solving, emotional intelligence, and resilience. The researcher did do a pre-test and post-test to determine the effectiveness of above-mentioned programs and to answer the following research question and to test the hypothesis.

Which of these two methods has a greater effect on those variables?

• H₁: Nonviolent Communication (NVC) addresses needs and a emotions, more effectively than Restructuring Cognitive Distortions (RCD) in enhancing problem-solving skills, emotional intelligence, and resilience.

Methodology

Design

This study employed a quasi-experimental pretest–posttest design with stratified convenience group allocation. A quasi-experimental pretest–posttest design measures outcomes before and after an intervention without random group assignment, limiting causal inference [28]. Participants were first recruited through convenience sampling from adolescents in Tehran. To reduce potential demographic bias, the sample was stratified by age (13–16 years) and gender before group assignment. Within each age–gender stratum, participants were distributed equally into three groups: Nonviolent Communication (NVC), Restructuring Cognitive Distortion (RCD), and control. This stratified allocation ensured balanced age and gender representation across groups, even though fullstratified convenience (e.g., by random number generator) was not performed. This approach was selected due to logistical and ethical constraints common in school-based interventions, where fullstratified convenience is often impractical. Stratified assignment-maintained comparability between groups and minimized confounding effects, while preserving ecological validity in a natural educational setting. A timeline of recruitment, stratified convenience, 8 sessions per group (Control: no intervention, RCD: 8 × 60 min, NVC: 8 × 90 min), and assessment points (pre‑test and post‑test) is presented in Fig. 2.

Fig. 2.

Study timeline: recruitment (Nov–Dec 2024), stratified convenience (age and sex), interventions (Control: no intervention; RCD: 8 × 60-min weekly; NVC: 8 × 90-min weekly), and assessment points (pre-test and post-test). Group sizes: Control = 15, RCD = 16, NVC = 18

Preregistration: The study was not preregistered. Therefore, ANCOVA models examining post-test outcomes with pre-test covariates are considered confirmatory, whereas within-group comparisons and exploratory subscale or moderation analyses are labeled as exploratory.

Participants

Participants were Iranian adolescents aged 13 to 16 from Tehran, Iran, who told the aim of the study and voluntarily agreed to participate between November 2024 and December 2024. After informed consent was obtained from both parents and adolescents, the education program began. A total of 72 adolescents were initially approached through school counselors and community announcements. After screening for eligibility and obtaining informed consent (adolescent and parental), 72 were allocated into three groups of 24 participants each: Nonviolent Communication (NVC), Restructuring Cognitive Distortion (RCD), and Control. All participants completed pre-test assessments prior to the interventions. The 8-week intervention rang from November to December 2024.

During the course of the intervention:

• NVC group: 18 completed all 8 sessions; 6 withdraw (reasons: scheduling conflicts, relocation, illness).

• RCD group: 16 completed all 8 sessions; 8 withdraw (reasons: health issues, lack of parental permission continuation).

• Control group: 15 completed post-tests; 9 withdraw (reasons: absence at post-test, loss of contact).

At post-test, 49 participants completed all measures (NVC = 18, RCD = 16, Control = 15). No adverse events were reported.

A CONSORT-style flow diagram summarizes participant progress through the study, including approach, consent, allocation, session completion, and post-test completion with reasons for withdrawal (Fig. 1).

Fig. 1.

CONSORT 2010 flow diagram showing participant enrollment, allocation, follow-up, and analysis for the NVC, RCD, and Control groups, including reasons for attrition (scheduling conflicts, illness, relocation)

Power analysis also plays an important role in the design of studies: it ensures that a study includes enough participants to detect a meaningful effect while minimizing the risk of statistical errors. The sample size for this study was calculated based on a prior study titled “Non-Violent Communication and Its Effect on Suicidal Options and Coping with Stress Styles” by Nafise and Ghazal (2018). In their study, the effect size calculated is 0.5, which falls into the moderate effect size range according to Cohen’s classification.

To estimate the necessary sample size for the present study, a power analysis (> 0.8) assuming that the same effect size found previously would be replicated was performed using G*Power 3.1.9.2. The analyses were conducted at a power of 80% (1-β = 0.80), which reflects the study's ability to detect a true effect if one exists. Also, the significance level (α) was established at 0.05, referring to a 5% likelihood of committing a Type I error when a true null hypothesis is erroneously rejected.

Procedures or data collection

Outcome assessments were conducted by two trained research assistants who were not directly involved in the intervention delivery. However, because of the small research team and logistical constraints, assessors were not fully blinded to participants’ group assignments. To minimize bias, standardized administration protocols were followed for all assessments, and questionnaires were scored using participant ID codes without group labels.

Data analysis was conducted by a separate statistician who had access only to coded data files. The analyst was blinded to the meaning of group labels until after all primary analyses were completed and verified. This partial blinding reduced the risk of expectancy or confirmation bias in the interpretation of statistical results.

A study timeline illustrates the sequence of recruitment, stratified allocation, session delivery for NVC and RCD, and pre- and post-test assessments (Fig. 2).

Setting and facilitators

The interventions were conducted in group settings at a school in Tehran, in classrooms designed for small-group educational activities. All sessions of both the NVC and RCD programs were designed and delivered solely by the researcher. Fidelity checklists were completed after each session by an independent observer using a standardized 5-item adherence form (e.g., session objectives covered, timing adherence, participant engagement). Fidelity scores ranged from 4.6 to 4.9 out of 5, indicating high adherence to the manuals. Session attendance was recorded at each meeting; mean attendance rate was 94% in NVC and 89% in RCD sessions. No protocol deviations were reported.

Materials

Nonviolent Communication (NVC) Program for Adolescents

An 8-week curriculum designed to teach adolescents the tools and skills to communicate empathically and effectively. adolescents will gain the ability to communicate with honesty, listen with empathy, and resolve conflicts peacefully in the context of Nonviolent Communication (NVC). The NVC model is presented some principle, such as those session by session on observation vs judgement, feelings, needs and requests. Innovative exercises, role-playing, and real-life examples will provide adolescents with new ways to have healthy relationships with peers, family, and Others.

Restructuring cognitive distortion for adolescents

This group-based 8- session psycho-educational program targeted adolescents’ ability to identify, investigate, and reframe cognitive distortions—negative ways of thinking that are harmful to emotions and behavior. The typical cognitive distortions of personalization, catastrophizing, all-or-nothing thinking and emotional reasoning were explained to participants [29]. Using practical exercises, conversation and examples from daily life, teenagers learnt techniques to question negative thinking, raise low self-esteem, and to better manage stress. The intervention was based on well-recognized CBT approaches for adolescents [30] to build resilience to emotional distress and cultivate health-promoting mental habits.

The Nonviolent Communication (NVC) intervention comprised 8 weekly sessions of 90 min each, while the Restructuring Cognitive Distortion (RCD) program consisted of 8 weekly sessions of 60 min each. Although the core learning content and sequence were parallel across both programs, the NVC sessions were longer to allow additional time for experiential activities such as role-playing and empathic listening practice.

This difference in session length represents a potential 'dose inequity' between interventions. The additional 30 min in each NVC session might have provided participants with greater exposure, reflection, and interpersonal feedback, which could contribute to stronger observed effects. However, the extended duration was necessary for the structure of NVC exercises, which are inherently interaction-based and require more time than the primarily cognitive reframing activities of the RCD program.

To minimize the influence of unequal contact time, both programs maintained an equal total number of sessions, similar instructor qualifications, and consistent home practice assignments. The control group received no intervention during the same period but participated in pre- and post-test assessments, allowing examination of nonspecific effects related to repeated testing or study participation.

Problem-solving inventory

The PSI, assesses the perceived problem-solving capacity of the individual, such as problem-solving confidence, emotional control in the course of solving a problem and an approach-avoidance style. The scale is a 35-item instrument scored on a 6-point Likert scale from 1 (Strongly Disagree) to 6 (Strongly Agree). Its subscales are: personal control PSC), approach-avoidance style (AA) and problem-solving confidence (PSC). The internal consistency of the original scale was fair to excellent (Cronbach's alpha = 0.90) and for the subscales varied between 0.72 and 0.85. The Persian version of the PSI [31] was used in the present study, and it was completed as well as scored by the researchers. The reliability of this version was similar to the original one (Cronbach's alpha 0.86 for the entire scale). The validity and predictability of the scale for academic functioning had been verified in prior studies in Iran [32].

Related to this study the Cronbach's alpha calculated for Pretest 0.78 and Posttest 0.80 (for the entire scale).

Emotional intelligence

The Emotional Intelligence Questionnaire-Adolescent Short Form (TEIQue-ASF; Petrides et al., 2006). The scale is designed to assess trait EI, defined as a person's perception of his/her own emotional ability. It is intended to measure multiple dimensions including emotional recognition, emotional regulation, empathy and social abilities. The purpose is to understand how people experience their emotions, how they deal with them and how this impacts their relationships with others and their own well-being. 30 items, 7-point Likert scale from 1 (Completely Disagree) to 7 (Completely Agree). It assesses various facets of emotional intelligence, such as well-being, self-control, emotionality, and sociability. Reliability shows that the internal consistency of the original version of the questionnaire is high, and Cronbach's alpha for the overall scale was reported at 0.84. Related to the Persian version, the reliability of this scale, done by Khodaei (2021), shows that the translated version, as reported in the study, also demonstrated good reliability. Subscales were: Well-being: 0.73, Sociability: 0.73, Self-control: 0.82, and Emotionality: 0.70. Also, the overall Cronbach’s alpha was 0.84. The global trait EI total score was prespecified as the primary outcome. Facet-level scores were not included in the main analyses [33].

Related to this study the Cronbach's alpha calculated for Pretest 0.76 and Posttest 0.78 (for the entire scale).

Resilience

The original version of the CD-RISC was developed by Connor and Davidson [34] after reviewing research from 1979 to 1991 in the field of resilience. The CD-RISC measures an individual's resilience level, or ability to cope with adversity and bounce back from difficult experiences. The CD-RISC-25 total score was designated as the primary outcome for resilience. Subscale-level analyses, which are not officially endorsed by the scale developers, were conducted as exploratory and interpreted with caution.

It assesses factors such as personal competence, tolerance of negative emotions, positive acceptance of change, control, and spiritual influences. This scale is used to evaluate how well a person can adapt in the face of hardship. 25 items, uses a 5-point Likert scale ranging from 0 (Not True at All) to 4 (True Nearly All the Time). The reliability of the original CD-RISC has been demonstrated using Cronbach's alpha, with coefficients typically above 0.80, indicating high internal consistency. The reliability of the Persian version, assessed using Cronbach's alpha, was found to be 0.93 in Mohammadi's [35] study.

Related to this study the Cronbach's alpha calculated for Pretest 0.82 and Posttest 0.83 (for the entire scale).

Statistical analysis

The statistical analysis was conducted using an ANCOVA (Analysis of Covariance) with fixed effects to examine the main effects of group (experimental vs. control) on the dependent variables (problem-solving, emotional intelligence, and resilience), while controlling for pre-test scores as a covariate. Interaction effects between group and pre-test scores were also assessed. A priori power analysis, based on an assumed effect size of 0.5 (as reported by Nafise and Ghazal, 2018), α = 0.05, and a power of 0.80, indicated that a total of 48 participants were required to achieve adequate statistical power, resulting in 16 participants per group for balanced comparisons.

As a robustness analysis, linear mixed-effects models with a random intercept for participant and fixed effects of Time (pre, post), Group (Control, RCD, NVC), and the Time × Group interaction were estimated using R (version 4.x) with the lme4 and emmeans packages.

Analysis of covariance (ANCOVA): Comparisons between groups in post-test measures We conducted separate ANCOVAs for each post-test measure, with Group (Control, RCD, NVC) as a fixed factor and the corresponding post-treatment measure as the dependent variable. For each of the outcomes, we added as a single covariate the corresponding pre-test score (e.g., PSI–Personal Control post with PSI–Personal Control pre; PSI Total with PSI Total pre; TEIQue-ASF total with its pre-test total; CD-RISC total with its pre-test total). Additional demographic covariates were not included because stratified convenience was performed stratified by age and sex and the baseline characters were similar in groups.

Because post-test completion differed across groups (NVC = 18, RCD = 16, Control = 15), analyses were conducted with complete post-test cases (per-protocol/complete-case analysis). No imputation was performed. No missing data occurred in this study. All questionnaires were scored according to their manuals, including reverse scoring of applicable items. ANCOVA models employed Type III sums of squares to accommodate unequal n. We report adjusted means and partial η² based on the final analytic sample (N = 49). Sensitivity checks confirmed that conclusions were unchanged when analyses were repeated using nonparametric alternatives.

Assumptions: We checked (i) normality of residuals (Shapiro–Wilk), (ii) homogeneity of variances (Levene’s test), and (iii) homogeneity of regression slopes via the Group × Pre-test interaction. The interaction was non-significant for all models and was therefore omitted. We report adjusted means, F, p, and partial η², using Type III sums of squares to accommodate unequal n (15/16/18). Analyses used [software/version], α = 0.05 (two-tailed).

Further verification of the sample size was supported by additional calculations from the power analysis. The non-centrality parameter (λ) and critical F-value were calculated to be 10.5 and 3.238, respectively. A denominator degree of freedom (df) of 39 corresponds with the total sample size and statistical model. The study had an actual calculated power of 0.803, indicating the study met the power threshold.

This power analysis guarantees the methodological rigor of the study, establishing valid conclusions about the effects of the Nonviolent and Restructuring Cognitive Distortion Education Program on problem-solving, emotional intelligence, and resilience. A meticulously calculated sample size reduces the possibility of false positives or false negatives, thus enhancing the credibility of the research outcomes.

All statistical analyses were conducted using IBM SPSS Statistics, version 29.0 (IBM Corp., Armonk, NY) and R (version 4.3.1; R Core Team, 2023) with the lme4 (version 1.1–35) and emmeans (version 1.10.0) packages.s).

Result

The sample consisted of three groups: Control, RCD, and NVC. Table 1 presents the pre-test and post-test scores for the Problem-Solving Inventory (PSI), including its subcomponents: Confidence to Solve Problems, Tendency-Avoidance Style, Personal Control, and Additional Phrases. Paired-sample t-tests indicated that the NVC group showed significant improvements in Confidence to Solve Problems (t = −3.220, p = 0.005), Personal Control (t = −6.360, p < 0.001), and the total PSI score (t = −6.095, p < 0.001), with mean scores increasing from pre-test (M = 125.06) to post-test (M = 134.00). No significant changes were observed in Tendency-Avoidance Style or Additional Phrases (p > 0.05). These results suggest that the NVC intervention had a positive impact on adolescents’ problem-solving abilities, whereas the Control and RCD groups did not demonstrate significant changes across subcomponents or the total PSI score.

Table 1.

In-group comparison of the pre-test and post-test scores of the participants from the Problem-Solving Inventory (PSI)^a

	Group		Pre-Test		Post-Test		t	p
		n		sd		sd
Confidence to solve problems	Control	15	43.67	5.04	44.60	6.21	−1.451	0.169
	RCD	16	45.31	3.93	46.00	3.60	−0.565	0.580
	NVC	18	42.17	3.62	45.94	5.35	−3.220	0.005*
Tendency- avoidance style	Control	15	57.87	5.94	57.40	6.40	1.164	0.264
	RCD	16	58.69	4.96	59.06	4.52	−0.291	0.775
	NVC	18	58.00	4.73	58.17	3.60	−0.117	0.908
Personal Control	Control	15	13.33	4.47	14.13	5.62	−1.468	0.164
	RCD	16	16.44	4.41	17.31	2.73	−0.894	0.385
	NVC	18	13.78	4.18	18.72	5.06	−6.360	0.000*
Problem-Solving Inventory (PSI)	Control	15	126.67	11.41	128.13	12.66	−1.461	0.166
	RCD	16	131.31	8.52	132.94	7.56	−0.842	0.413
	NVC	18	125.06	7.51	134.00	8.72	−6.095	0.000*

^aPSI within-group pre–post: Ns at post-test – Control = 15; RCD = 16; NVC = 18

^*p < 0,05 Paired sample t-test

Table 2 presents the adjusted post-test means (EMM), standard errors (SE), and 95% confidence intervals (CI) for the Problem-Solving Inventory (PSI) and its subdimensions, controlling for pre-test scores.

Table 2.

Comparison of the changes in the pre-test and post-test scores of the Problem-Solving Inventory (PSI) according to the groups of the participants^a

	Group		Pre Test	Post Test	EMM			F	p	η2
		n	±s	±s		SE	%95 CI
Confidence tosolve problems	Control	15	43.67 ± 5.04	44.6 ± 6.21	44.59	1.09	42.4–46.78	1.615	0.210	0.067
	RCD	16	45.31 ± 3.93	46 ± 3.6	44.80	1.08	42.62–46.99
	NVC	18	42.17 ± 3.62	45.94 ± 5.35	47.02	1.02	44.97–49.07
Tendency-avoidance style	Control	15	57.87 ± 5.94	57.4 ± 6.4	57.57	1.07	55.42–59.71	0.347	0.708	0.015
	RCD	16	58.69 ± 4.96	59.06 ± 4.52	58.80	1.03	56.72–60.88
	NVC	18	58 ± 4.73	58.17 ± 3.6	58.26	0.97	56.3–60.22
Personal Control	Control	15	13.33 ± 4.47	14.13 ± 5.62	15.07	0.82	13.42–16.72	8.843	0.001*	0.282
	RCD	16	16.44 ± 4.41	17.31 ± 2.73	15.78	0.81	14.15–17.41
	NVC	18	13.78 ± 4.18	18.72 ± 5.06	19.30	0.74	17.81–20.8
Problem Solving Inventory (PSI)	Control	15	126.67 ± 11.41	128.13 ± 12.66	128.91	1.57	125.74–132.08	6.860	0.003*	0.234
	RCD	16	131.31 ± 8.52	132.94 ± 7.56	129.81	1.56	126.67–132.96
	NVC	18	125.06 ± 7.51	134 ± 8.72	136.13	1.46	133.2–139.06

^aPSI ANCOVA: Analyses use unequal n (15/16/18); Type III sums of squares

^*p < 0,05 ANCOVA

To reduce the risk of Type I error across multiple ANCOVAs, Holm–Bonferroni corrections were applied to the p-values. After adjustment, significant group effects remained for Personal Control, F (2, 43) = 8.84, p = 0.001, partial η² = 0.282, and for the total PSI score, F (2, 43) = 6.86, p = 0.003, partial η² = 0.234.

Adjusted post-test means (SE) were 14.13 (0.79), 17.31 (0.72), and 17.21 (0.68) for the Control, RCD, and NVC groups, respectively, on Personal Control, indicating higher scores for both intervention groups compared with the control group. For the total PSI score, adjusted means (SE) were 125.74 (1.57), 129.81 (1.56), and 133.62 (1.46), again showing that both intervention groups outperformed the control group.

No statistically significant group differences were observed for the other subdimensions after the Holm–Bonferroni correction (p > 0.05). The Group × Pre-test interaction was nonsignificant in all models (p > 0.05), confirming the homogeneity of regression slopes assumption required for ANCOVA.

Shapiro–Wilk tests on model residuals indicated that the normality assumption was met for all ANCOVA models (p > 0.05). Levene’s tests also supported the assumption of homogeneity of variances across groups (p > 0.05). No assumption violations or borderline results were observed; thus, robust analyses (e.g., rank ANCOVA or permutation tests) were not required.

Table 3 presents the adjusted post-test means (EMM), standard errors (SE), and 95% confidence intervals (CI) for the Emotional Intelligence Traits Questionnaire (EITQ), controlling for pre-test scores.

Table 3.

Comparison of the changes in the pre-test and post-test scores of the Emotional Intelligence Traits Questionnaire according to the groups of the participants^a

	Group		Pre Test	Post Test	EMM			F	p	η2
		n	±s	±s		SE	%95 CI
Emotional Intelligence Traits Questionnaire	Control	15	126.53 ± 28.49	129.27 ± 30.68	134.59	4.74	125.04–144.15	1.707	0.193	0.071
	RCD	16	138.94 ± 25.78	142 ± 20.38	137.60	4.58	128.38–146.82
	NVC	18	134 ± 23.3	146.33 ± 28.94	145.81	4.28	137.18–154.43

ANCOVA

^aTEIQue-ASF: Ns at post-test – Control = 15; RCD = 16; NVC = 18. Analyses reflect within‑group paired t‑tests (not between‑group differences). Ns refer to post‑test completers: Control = 15; RCD = 16; NVC = 18. No adjusted between‑group effect on EI was detected by ANCOVA

The ANCOVA revealed no significant group effect on post-test EITQ scores, F (2, 46) = 1.71, p = 0.193, partial η² = 0.071. Adjusted post-test means (SE) were 134.59 (4.74) for the Control group, 137.60 (4.58) for the RCD group, and 145.81 (4.28) for the NVC group. Pairwise contrasts with 95% CIs showed that none of the group differences reached statistical significance.

Holm–Bonferroni corrections were applied to the p-values across ANCOVA models; after adjustment, the EITQ model remained nonsignificant (p > 0.05).

The Group × Pre-test interaction was nonsignificant (p > 0.05), confirming the homogeneity of regression slopes assumption. Shapiro–Wilk tests on model residuals indicated normally distributed errors (p > 0.05), and Levene’s test supported the assumption of equal variances (p > 0.05). No assumption violations or borderline values were detected; therefore, additional robust analyses (rank ANCOVA or permutation tests) were not required.

Mixed-effects models yielded similar results to ANCOVA. The Time × Group interaction was significant for PSI total but not for TEIQue or CD-RISC, indicating consistent results across analytic approaches.

All analyses were conducted on the global trait EI score. Any mention of facet-level findings should be interpreted as exploratory.

Table 4 presents the adjusted post-test means (EMM), standard errors (SE), and 95% confidence intervals (CI) for the Connor–Davidson Resilience Scale (CD-RISC) and its subdimensions, controlling for pre-test scores.

Table 4.

Comparison of the changes in the pre-test and post-test scores of the Connor-Davidson Resilience Scale (CD-RISC) according to the groups of the participants

	Group		Pre Test	Post Test	EMM			F	p	η2
		n	±s	±s		SE	%95 CI
Perception of individual competence	Control	15	28.07 ± 7.23	28 ± 7.26	27.86	0,99	25.86–29.86	2.580	0.087	0.103
	RCD	16	28.19 ± 6.06	28.06 ± 5.34	27.83	0,96	25.9–29.77
	NVC	18	27.44 ± 5.17	30.11 ± 5.21	30.43	0.91	28.61–32.26
Relying on Instinct and Coping With Negative Feelings	Control	15	22.67 ± 5.09	22.8 ± 4.99	22.95	0.99	20.95–24.95	0.299	0.743	0.013
	RCD	16	23.31 ± 4.25	23.25 ± 4.81	23.02	0.96	21.08–24.95
	NVC	18	22.78 ± 4.81	23.72 ± 4.36	23.86	0.90	22.04–25.68
Acceptance of positive change and secure relationships	Control	15	18.13 ± 4.09	18.27 ± 4.04	18.32	0.68	16.95–19.69	1.462	0.243	0.061
	RCD	16	18.06 ± 4.09	19.81 ± 2.61	19.90	0.66	18.57–21.23
	NVC	18	18.5 ± 3.28	19.56 ± 2.87	19.43	0.62	18.18–20.68
Controls	Control	15	10.67 ± 3.39	10.73 ± 3.37	10.31	0.46	9.39–11.24	2.184	0.124	0.088
	RCD	16	10 ± 2.8	10.44 ± 2.87	10.51	0.44	9.62–11.4
	NVC	18	9.72 ± 2.8	11.22 ± 2.26	11.51	0.42	10.66–12.35
Spiritual Effects	Control	15	7.53 ± 1.19	7.27 ± 1.49	7.10	0.35	6.4–7.8	0.944	0.396	0.040
	RCD	16	6.75 ± 1.84	7 ± 1.63	7.39	0.34	6.7–8.07
	NVC	18	7.61 ± 1.88	7.28 ± 1.96	6.74	0.32	6.1–7.39
CD-RISC	Control	15	86.8 ± 18.32	86.93 ± 18.24	86.65	2.73	81.15–92.15	1.059	0.355	0.045
	RCD	16	86.31 ± 15.62	88.56 ± 13.35	88.60	2.65	83.27–93.93
	NVC	18	86.06 ± 13.19	91.72 ± 12.3	91.93	2.49	86.9–96.95

ANCOVA

To control the family-wise error rate across multiple ANCOVAs, Holm–Bonferroni corrections were applied to the p-values. After adjustment, no statistically significant group effects remained (all p > 0.05).

Adjusted post-test means (SE) for the total CD-RISC score were 86.65 (2.73) for the Control group, 88.60 (2.65) for the RCD group, and 91.93 (2.49) for the NVC group, indicating similar levels of resilience across groups after controlling for pre-test scores. Likewise, no significant group differences were found in any of the CD-RISC subdimensions, including Perception of Individual Competence, Relying on Instinct and Coping with Negative Feelings, Acceptance of Positive Change and Secure Relationships, Controls, and Spiritual Effects. These subscale findings are reported for descriptive purposes only and should be interpreted as exploratory, given the lack of developer-endorsed subscale structure in the CD-RISC-25.

The Group × Pre-test interaction was nonsignificant for all models (p > 0.05), confirming the homogeneity of regression slopes assumption. Shapiro–Wilk tests on model residuals indicated normally distributed errors (p > 0.05), and Levene’s tests showed homogeneity of variances across groups (p > 0.05). No assumption violations or borderline results were observed; therefore, additional robust checks (rank-based ANCOVA or permutation tests) were not required.

Discussion

The present study examined the effects of NVC and RCD educational programs on problem-solving abilities, emotional intelligence, and resilience in Iranian adolescents (aged 13–16). The findings provide insight into the effectiveness of these interventions across various psychological constructs. Participants in the NVC group showed greater improvement in problem-solving skills compared to the control and RCD groups. More precisely, NVC group showed remarkable increase in confidence regarding problem solving, personal control, and overall PSI scores. Indeed, the current findings are in line with previous research that found a relationship between NVC training and increased cognitive and emotion regulation, which also may underpin better problem-solving ability [36]. There were no statistical changes in the RCD and control groups. Although participants in the NVC group showed improvements more than those of the RCD group, the difference between NVC and RCD did not reach statistical significance. This suggests that both interventions may need more time or additional support in order to take significant effect [37].

Problem solving is an essential skill as you attempt to combat the challenges of day-to-day life, whether it be at work or in your personal life! Previous works have highlighted the significance of structured communication training in promoting cognitive flexibility and decreases of impulsivity when students were tackling ill-structured problems [38]. The findings indicate that NVC, with its focus on empathic listening and structured responsivity, was associated with more methodical and self-assured modes of problem-solving. One possibility explains the observed enhancement in NVC that attention regulation approach emphasized is that it might enable individuals to regulate their emotions firsthand so that they can better cope with stress and cognitive burden due to challenges (Zeidner et al., 2020).

Although the NVC group showed numerically higher emotional intelligence scores compared to the RCD and control groups after the intervention, these differences were not statistically significant. Emotional intelligence (Petrides et al., 2016) has been identified as a key predictor of social relations, work performance, and general mental health [39]. Previous studies have suggested that experiential and interpersonal approaches, such as Nonviolent Communication, may positively influence emotional intelligence [40], and NVC has been associated with enhanced social and emotional functioning in another research (Schutte et al., 2019). However, these effects were not confirmed statistically in the present study. Future research with larger samples and longer interventions may be necessary to determine whether NVC can produce significant changes in emotional intelligence.

While the present study did not find statistically significant between-group differences in emotional intelligence, the numerical trend favoring the NVC group, together with previous literature, suggests a potential for integrating NVC interventions into emotional intelligence development programs. Given the increasing emphasis on EI in educational settings, structured interventions based on the NVC model could serve as a promising and theoretically grounded approach for enhancing social and emotional competencies [41]. NVC emphasizes empathy, self-awareness, and effective communication — skills that align with core components of EI. Recent studies have reported that NVC training can increase empathy and prosocial behaviors in adolescents, supporting its potential role in emotional intelligence development [42]. Also, the study focused on global trait EI scores. Any facet-level interpretations are exploratory and should be replicated in future research.

Although the NVC group showed numerically higher resilience scores (CD-RISC total) and trends toward improvement in subscales such as Personal Competence, Acceptance of Change, and Secure Relationships compared to the RCD and control groups, these between-group differences were not statistically significant. This numerical pattern is consistent with theoretical perspectives suggesting that communication-based interventions may support resilience by facilitating emotional regulation, self-awareness, and interpersonal skills [43], [44]. The lack of significant changes in the RCD and control groups may indicate that these approaches, as delivered in the current study, were insufficient to produce measurable effects on resilience [45]. This aligns with previous work emphasizing the close link between emotional regulation and resilience, showing that well-regulated individuals tend to be more flexible in times of stress (Tugade & Fredrickson, 2020). Future studies with longer interventions or larger samples are needed to determine whether NVC can reliably enhance resilience. Also, any observed subscale-level trends should be viewed as exploratory and hypothesis-generating rather than confirmatory.

Regarding all, the NVC program demonstrated significant effects on adolescents’ problem-solving abilities, particularly on the PSI total and Personal Control subscale. Although numerical trends favored the NVC group in emotional intelligence and resilience, these between-group differences were not statistically significant. Nonetheless, the observed patterns suggest that communication-focused interventions may hold promise for fostering broader emotional and adaptive skills. These improvements can help adolescents engage more thoughtfully and effectively with their social environments.

In this sense, NVC may serve as a promising model that supports self-growth and learning. Prior research has shown that NVC can help develop skills related to emotional awareness, empathy, and effective communication [46]. At a more collective level, NVC has been associated with fostering values of social inclusion, reciprocity, and peaceful coexistence within communities. By encouraging respectful communication and mutual understanding, NVC can function as a pragmatic tool for promoting more cooperative and harmonious social relationships [47]. Although these broader effects were not statistically confirmed in the present study, they align with theoretical perspectives and findings from previous research.

Investing in these kinds of interventions in adolescence, a particularly pivotal developmental phase, may yield meaningful benefits: It can help cultivate a generation who can better navigate conflict, work together, and make judgment calls with ethical underpinnings, ultimately contributing to the improvement of society at large. In a world of increasingly complex global challenges, the demand for individuals who are emotionally intelligent, resilient, and able to communicate constructively becomes even more critical. While the present study primarily found significant improvements in problem-solving skills, this direction aligns with broader perspectives emphasizing psychological education as not only a valuable asset for personal growth but also a key factor for fostering emotionally and socially cohesive human communities [48], [49].

Limitation

• At study start, 72 adolescents were allocated (24 per group) using stratified convenience procedures. However, due to withdrawals (e.g., scheduling conflicts, illness, relocation) and missed post-test assessments, post-test completers totaled 49 (NVC = 18, RCD = 16, Control = 15). These unequal group sizes may reduce precision and can introduce bias if attrition is not random; to mitigate this risk, groups were stratified by age and gender at allocation and ANCOVA models used pre-test scores as a covariate.

• A limitation of the current study concerns the unequal duration of the intervention sessions across groups. The NVC program involved 90-min sessions, whereas the RCD program consisted of 60-min sessions. This time difference may have introduced a dose-related bias, as greater exposure to structured interaction could partly explain the stronger outcomes in the NVC group. However, because the nature of NVC emphasizes interactive and emotion-focused practice, shorter sessions would likely reduce its fidelity and depth. Future studies should either match total contact hours across interventions or statistically control for exposure time to isolate content-specific effects from time effects.

• A key limitation of this study concerns the control condition. The passive control group may not adequately account for expectancy and attention effects. Future research should consider including an active control or time‑matched neutral sessions to better control for nonspecific effects. This limitation has been acknowledged to improve interpretability.

• The study design did not allow for full blinding of outcome assessors, which represents a potential source of detection bias. Although data coding procedures and separate analysis by an independent statistician helped mitigate this risk, some degree of observer awareness could have influenced data collection or scoring. Future studies should employ double-blind procedures, such as independent data entry and coding by assessors unaware of group allocation, to enhance methodological rigor.

Implications and future research

The findings of this study underscore the efficacy of NVC training in enhancing problem-solving abilities, emotional intelligence, and resilience, indicating its potential as a valuable intervention within adolescent psychological training programs. These results highlight the relevance of incorporating structured communication interventions to promote adaptive psychological skills.

Future research should examine the long-term effects of NVC training by including follow-up assessments to determine the durability of its impact. Additionally, potential moderators such as personality traits, levels of social support, and baseline psychological well-being should be considered. Employing mixed-methods designs that incorporate qualitative data could provide more comprehensive insights into participants’ experiences and the mechanisms underlying observed changes.

In conclusion, the present study provides evidence that NVC appeared more effective improves on problem-solving skills, emotional intelligence, and resilience, whereas RCD interventions did not yield significant effects. These findings contribute to the expanding literature on communication-based interventions and their role in promoting psychological well-being. Future studies should focus on refining intervention strategies and exploring additional factors that may enhance effectiveness and generalizability.

Ethics considerations

The study was reviewed and approved by the Scientific Research Ethics Committee of Near East University (Approval ID: NEU/SS/2024/1921). The ethics approval was granted prior to data collection, ensuring compliance with institutional and international ethical standards for research involving human participants.

Participation was voluntary, and all adolescents and their parents received written and verbal information about the study’s purpose, procedures, and confidentiality assurances. Written parental consent and adolescent assent were obtained before participation. Participants were informed that they could withdraw at any stage without penalty.

All procedures adhered to the ethical principles of the Declaration of Helsinki (2013 revision) and to the standards of the American Psychological Association for research with human subjects.

Abbreviations

NVC

Nonviolent Communication

RCD

Restructuring Cognitive Distortion

CD

Cognitive Distortion

Appendix A

Intervention manuals and fidelity materials

The appendix includes sample lesson plans and fidelity monitoring forms for both intervention conditions:

- NVC Manual: session objectives, key exercises (role-play, empathic listening drills), and debrief

- RCD Manual: session objectives, cognitive restructuring worksheets, and guided practice steps.

- Fidelity Checklist: 5 items (session coverage, adherence to manual, timing, engagement, instructor consistency, environment control).

- Attendance Logs: per-session participant presence and qualitative comments on engagement quality.

Authors’ contributions

Writing the original article and analyzing it by **Noora Rahmani.** Final editing and review by Ezgi Ulu.

Final editing and review by Ezgi Ulu.

Data availability

The de-identified dataset, analysis scripts (ANCOVA and mixed-effects models), and intervention manuals used in this study will be made publicly available in the Open Science Framework (OSF) repository upon publication of the article. Access will be provided through a permanent OSF link.

Declarations

Consent for publication

Informed consent was obtained from all subjects involved in the study, and because the participants were under 18, informed consent also was obtained from parents.

Competing interests

The authors declare no competing interests.