Malevolent Creativity Behavior Scale-Brazilian Portuguese: Cross-Cultural Adaptation and Psychometric Properties
ABSTRACT
This study aimed to adapt the Malevolent Creativity Behavior Scale (MCBS) to Brazilian Portuguese and to estimate validity evidence based on the internal structure and relationships with other variables and reliability. The sample consisted of 382 Brazilians aged between 18 and 71 (M = 32.18, SD = 12.89), of which 68.06% were female. Through the content validity coefficient, the judges' analysis indicated the adequacy of the items. In addition, confirmatory factor analysis demonstrated the adequacy of three factors, all with good reliability indicators. The convergent validity of the MCBS with the SD4 and the PSCD indicated that all correlations were positive, involving Machiavellianism, sadism, and psychopathy. Men presented significantly higher levels of all three malevolent creativity when compared to women. Thus, the results suggest that the MBCS adequately measures malevolent creativity in the Brazilian population.
1 Introduction
Creativity refers to the ability to generate original, new, and useful ideas within a context (Runco and Jaeger 2012). This quality is often associated with positive outcomes (Oliveira et al. 2016), including personal and professional fulfillment, academic performance, and mental health. It is a valuable characteristic that contributes to one's personal development and quality of life (Jia et al. 2020). According to most theories on creativity, this characteristic is regarded as a positive force that benefits individuals and society as a whole, emphasizing its positive attributes aimed at desirable and constructive purposes (Kappor and Kaufman 2022). However, the realization that creativity can also be a negative characteristic has taken hold in scientific research (Reis et al. 2024). Creativity has recently been investigated within the context of a valence perspective in order to recognize its potential application for positive and negative purposes (Hao et al. 2020). It involves the use of so-called malevolent creativity for harmful, unethical, and malevolent purposes (Xu et al. 2023), being the dark side of creativity (Bedu-Addo et al. 2023).
It refers to the use of this potential and the original ideas to harm individuals or society (Kappor 2019) or for unfair gains or advantages (Palmer et al. 2020) in a deliberately intended or unintended by-product of the creative action (Reis et al. 2024). This definition of malevolent creativity includes products that can be beneficial to their creators but harmful to others (Perchtold-Stefan, Rominger, and Fink 2023).
There has been significant progress and an increasing amount of research regarding the malevolent expression of creativity, focusing on individual, situational, and societal factors that can influence it (Kappor and Kaufman 2022; Kapoor and Khan 2016; Mitchell and Reiter-Palmon 2022; Reis et al. 2024). Nonetheless, this is still a field that has not been explored much. Although the concept of malevolent creativity is typically associated with individuals who have carried out far-reaching harmful acts, it is essential to clarify that all individuals have the potential to express creativity in its malevolent form, and it is not exclusive to criminals and terrorists (Jia et al. 2020).
It has been found that most research on malevolent creativity is focused on the stressors affecting individuals and their environments in order to identify their main determinants (Kappor and Khan 2017). According to previous studies, malevolent creativity can be easily stimulated in the general population when people are presented with situations that provoke anger or represent unfair behavior from peers and instructed to come up with creative ideas to avenge them (Perchtold-Stefan et al. 2023). The identification of predictors of malevolent creative behavior, as well as their interaction mechanisms, has been a focus of researchers seeking to understand the nature of this type of creative expression (Bedu-Addo et al. 2023; Li et al. 2022) due to its potential to cause unanticipated harm and future losses (Kappor 2023).
Knowledge about creativity's malevolent side can reduce opportunities for these to emerge. More specifically, research shows that some personality traits, such as subclinical psychopathy, play a significant role in explaining malevolent creativity (Hao et al. 2020; Perchtold-Stefan et al. 2023). Maladaptive personality characteristics like self-centeredness, manipulation, lower inhibition, antisocial traits, sadism, lack of empathy, lower conscientiousness, grandiosity, manipulativeness, and callousness have been associated with malevolent creativity (Batey et al. 2022; Kappor 2023; Kapoor et al. 2024; Lee and Dow 2011; Perchtold-Stefan, Fink, et al. 2021; Zhou et al. 2024). Some studies have shown that conduct disorders, including antisocial tendencies like aggressive behavior and anger, have a significant effect on malevolent creativity (Cheng et al. 2021; Jiaq et al. 2024; Perchtold-Stefan, Rominger, et al. 2021; Perchtold-Stefan et al., 2023; Perchtold-Stefan, Rominger, and Fink 2023; Rui et al. 2021; Wang et al. 2024).
Several personality traits, particularly those associated with the Dark Triad of personality (i.e., Machiavellianism, narcissism, and psychopathy), can contribute to the manifestation of malevolent creativity (Batey et al. 2022; Gao, Cheng, et al. 2022; Kappor and Khan 2017; Lebuda et al. 2021; Perchtold-Stefan et al. 2020; Szabó et al. 2022). A meta-analysis conducted by Zhou et al. (2024) suggested that Dark Triad traits were crucial to the induction of malevolent creativity. In this sense, evidence suggests that the Dark Triad can facilitate the development of malevolent creativity (Geng et al. 2024).
Because the concept of malevolent creativity has become increasingly popular, several instruments have been developed for assessing it (Waldie et al. 2021). This implies that measuring malevolent creativity becomes vital to operationalizing the theoretical foundations of this type of creativity (Cropley and Cropley 2019). There is also a focus on examining gender differences in malevolent creativity (Al-Mahdawi et al. 2022; Kappor 2019; Perchtold-Stefan et al. 2023).
One widely used scale is the Malevolent Creativity Behavior Scale (MCBS), developed by Hao et al. (2016). It was developed based on 13 malevolent creative behaviors that may occur in everyday life (e.g., deceptions, tricks, lies, betrayals, revenge, rumormongering, etc.). Respondents are asked to rate each behavior regarding how frequently they occur (0 = never, 1 = few times, 2 = sometimes, 3 = often, and 4 = usually). The items are grouped into three factors. Hurting people (factor 1) involves the application of original ideas to purposely harm others or self, often to gain an unfair advantage through manipulation, threat, or harm. Lying (factor 2) includes flexibility and divergent thinking to be dishonest. Playing tricks (factor 3) requires creative thinking to invent new tricks and update them frequently. Higher scores indicate a greater frequency of malevolently creative behaviors.
Considering the importance of research on malevolent creativity, particularly concerning its measurement, a process of adapting the MCBS for use in Brazil has begun (Bonfá-Araujo et al. 2023). Several studies using this instrument have already been published in different samples and countries (Al-Mahdawi et al. 2022; Bedu-Addo et al. 2023; Dan et al. 2021; Gao, Qiao, et al. 2022; Hao et al. 2020; Jia et al. 2020; Malik et al. 2020; Meshkowa et al. 2022; Osman et al. 2024; Perchtold-Stefan et al. 2023; Szabó et al. 2022; Xu et al. 2021; Xu et al. 2023; Li et al. 2022; Zhao et al. 2022). Thus, we sought to adapt the Malevolent Creativity Behavior Scale and test its association with dark and antisocial traits while also considering the measure's adequacy.
2 Method
2.1 Participants and Procedures
We collected data from 382 Brazilians aged 18 to 71 (M = 32.18, SD = 12.89). Participants were primarily women (68.06%), single (52.35%), or married (18.06%), living in the capital or their respective state (33.50%) or in a medium-sized city between 100 thousand and 500 thousand inhabitants (25.13%), in the Southeast (48.95%) or South (16.75%) region of Brazil.
To perform the translation and adaptation process of the Malevolent Creativity Behavior Scale (MCBS; Hao et al. 2016) to Brazilian Portuguese, we followed the recommendations made by the International Test Commission (2017), consisting of translation, evaluation of the quality of the translation, back-translation, and pilot study. We asked two Ph.D. with an emphasis on psychological assessment and English and Brazilian Portuguese proficiency to translate the original items, considering the particularities between both languages. After synthesizing both translations into one, the items were back-translated and tested with a group of 10 people for adequacy (Bonfá-Araujo et al. 2023). The complete project was developed according to the Declaration of Helsinki and its later amendments, and it was approved by an Ethical Board Committee. To be a part of the research, participants had to voluntarily accept the information in the consent form. The instruments were shown in the order presented in the measures section.
2.2 Measures
2.2.1 Malevolent Creativity Behavior Scale (MCBS; Hao et al. 2016)
To measure malevolent creativity, we adapted the MCBS to Brazilian Portuguese, which is designed to assess hurting people (How often do you think about ideas to take revenge when being unfairly treated?), lying (How often do you fabricate lies to simplify a problem situation), and playing tricks (How often do you have ideas about how to pull pranks on others?). The measure has 13 items, six for hurting people, four for lying, and three for playing tricks, and it is answered on a 5-point Likert scale (0 = Never to 4 = Usually). Reliability is shown in the results section.
2.2.2 Short Dark Tetrad (SD4; Paulhus et al. 2021)
To measure dark personality traits, we used the SD4 adapted to Brazilian Portuguese (Guilhermino et al. 2025), which is designed to assess Machiavellianism (It's not wise to let people know your secrets), narcissism (People see me as a natural leader), psychopathy (People often say I'm out of control) and sadism (Watching a fistfight excites me). The measure has 28 items, seven per dimension, and it is answered on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree). For the present study, reliability was considered adequate: Machiavellianism (α = 0.70), narcissism (α = 0.73), psychopathy (α = 0.80), and sadism (α = 0.84).
2.2.3 Proposed Specifiers for Conduct Disorder (PSCD; Salekin and Hare 2016)
To measure antisocial traits, we used the PSCD adapted to Brazilian Portuguese (Pechorro et al. 2025), which is designed to assess interpersonal (Grandiose-Manipulative, I take advantage of others), affective (Callous-Unemotional, I rarely feel guilty or remorse), lifestyle (Daring-Impulsive, I feel like I need a lot of stimulation), and Conduct Disorder symptoms (I have stolen things). The measure has 24 items, six per dimension, and it is answered on a 3-point Likert-type scale (0 = not true, 1 = somewhat true, and 2 = true). For the present study, reliability was considered mostly adequate: Grandiose–Manipulative (α = 0.66), Callous–Unemotional (α = 0.79), Daring–Impulsive (α = 0.80), and conduct disorder (α = 0.77).
2.3 Data Analysis
We first tested descriptive statistics for the MCBS. Next, we conducted a Confirmatory Factor Analysis (CFA) to examine two possible models, a unidimensional factor and a 4-factor solution, using the diagonally weighted least squares (DWLS) estimator. We relied on commonly used indexes: Comparative Fit Index (CFI, above 0.90), Tucker-Lewis Index (TLI, above 0.90), Root Mean Square Error of Approximation (RMSEA, below 0.08), and Standardized Root Mean Squared Residual (SRMR, below 0.08). We also estimated the reliability of the psychometric measures using Cronbach's alpha (α), McDonald's omega (ω; both if ≤ 0.59 low, ≥ 0.60 marginal, ≥ 0.70 fair/acceptable, ≥ 0.80 good, and ≥ 0.90 excellent). Moreover, we finallytested Pearson's correlation and Student's t-test when appropriate. All analyses were performed in JASP (JASP Team 2024).
3 Results
Table 1 shows the descriptive information of the MCBS items, considering the total sample. Next, we tested the two confirmatory models (i.e., 1-factor and 3-factor solution), presented in Table 2. Results suggest that the 3-factor solution, proposed initially by Hao et al. (2016), is also the one with the best adequacy for the Brazilian sample, considering the total sample or sample by gender.
Mean | Standard deviation | Skewness | Kurtosis | Minimum | Maximum | |
---|---|---|---|---|---|---|
MCBS1 | 1.61 | 1.28 | 0.54 | −0.76 | 0 | 4 |
MCBS2 | 1.07 | 1.28 | 1.02 | −0.10 | 0 | 4 |
MCBS3 | 0.85 | 1.20 | 1.33 | 0.68 | 0 | 4 |
MCBS4 | 0.50 | 0.94 | 2.09 | 3.75 | 0 | 4 |
MCBS5 | 0.96 | 1.30 | 1.17 | 0.11 | 0 | 4 |
MCBS6 | 0.61 | 1.15 | 1.87 | 2.31 | 0 | 4 |
MCBS7 | 1.09 | 1.23 | 0.98 | −0.02 | 0 | 4 |
MCBS8 | 1.54 | 1.24 | 0.51 | −0.64 | 0 | 4 |
MCBS9 | 0.76 | 1.09 | 1.46 | 1.31 | 0 | 4 |
MCBS10 | 1.40 | 1.31 | 0.71 | −0.64 | 0 | 4 |
MCBS11 | 0.58 | 1.06 | 1.92 | 2.79 | 0 | 4 |
MCBS12 | 0.35 | 0.85 | 2.78 | 7.51 | 0 | 4 |
MCBS13 | 1.12 | 1.33 | 0.95 | −0.35 | 0 | 4 |
χ (df) | CFI | TLI | RMSEA (90% CI) | SRMR | |
---|---|---|---|---|---|
Total sample | |||||
1-factor | 128.437 (65) | 0.98 | 0.97 | 0.05 (0.03 to 0.06) | 0.08 |
3-factor | 68.162 (62) | 0.99 | 0.99 | 0.01 (0.00 to 0.03) | 0.06 |
Male sample | |||||
1-factor | 95.824 (65) | 0.97 | 0.97 | 0.06 (0.03 to 0.08) | 0.1 |
3-factor | 61.707 (62) | 1.00 | 1.00 | 0.00 (0.00 to 0.00) | 0.08 |
Female sample | |||||
1-factor | 73.102 (65) | 0.99 | 0.99 | 0.02 (0.00 to 0.04) | 0.09 |
3-factor | 37.614 (62) | 1.00 | 1.00 | 0.00 (0.00 to 0.00) | 0.06 |
Table 3 shows the standardized factor loadings of the 3-factor solution. The loading values were above the recommended 0.30 cutoff, ranging between 0.48 and 0.82. In Table 4, we present the intercorrelation between the three MCBS factors and reliability. All three factors presented good fits, higher than 0.72. Considering the reliability, we can see that correlations between factors are positive and significant.
Items | Loadings |
---|---|
How often do you think about ideas to take revenge when being unfairly treated? Com que frequência você pensa em ideias para se vingar ao ser tratado injustamente? |
0.76 |
How often do you have ideas about new ways to punish people? Com que frequência você tem ideias sobre novas maneiras de punir as pessoas? |
0.79 |
How often do you have ideas about how to suppress people who are in your way? Com que frequência você tem ideias sobre como reprimir as pessoas que estão no seu caminho? |
0.81 |
How often do you engage in an original form of sabotage? Com que frequência você se envolve em uma forma original de sabotagem? |
0.61 |
How often do you have ideas to hurt yourself? Com que frequência você tem ideias para se machucar? |
0.48 |
How often do you think about the strategies of hurting others in the rough world? Com que frequência você pensa nas estratégias de ferir os outros no mundo difícil? |
0.79 |
How often do you fabricate lies to simplify a problem situation? Com que frequência você fabrica mentiras para simplificar uma situação-problema? |
0.82 |
How often do you think about excuses to justify your wrongdoings? Com que frequência você pensa em desculpas para justificar seus erros? |
0.74 |
How often do you tell lies without worrying about being nailed? Com que frequência você conta mentiras sem se preocupar em ser pego? |
0.74 |
How often do you think of ways to conceal your misdoings from others? Com que frequência você pensa em maneiras de esconder seus erros dos outros? |
0.79 |
How often do you have ideas about how to pull pranks on others? Com que frequência você tem ideias sobre como pregar peças nos outros? |
0.66 |
How often do you play tricks on people as revenge? Com que frequência você prega peças nas pessoas como vingança? |
0.67 |
How often do you think of ideas on the margins of rules, when conventional ways do not work? Com que frequência você pensa em ideias à margem das regras, quando as formas convencionais não funcionam? |
0.73 |
Hurting people | Lying | Playing tricks | |
---|---|---|---|
Hurting people | |||
Lying | 0.64* | ||
Playing tricks | 0.70* | 0.60* | |
Cronbach's alpha | 0.85 | 0.85 | 0.73 |
McDonald's omega | 0.86 | 0.85 | 0.72 |
- * p ≤ 0.001.
Finally, in Tables 5 and 6, we present the convergent validity for the MCBS with the SD4 and the PSCD, as well as the mean differences between males and females. All correlations were positive, with Machiavellianism, sadism, and callous-unemotional traits showing the highest values when correlated with the three factors from the MCBS. Furthermore, males showed significantly higher levels of all three MCBS factors than females.
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | |
---|---|---|---|---|---|---|---|---|---|---|---|
1. MCBS hurting people | |||||||||||
2. MCBS lying | 0.64* | ||||||||||
3. MCBS playing tricks | 0.70* | 0.60* | |||||||||
4. MCSB total score | 0.92* | 0.85* | 0.82* | ||||||||
5. SD4 machiavellianism | 0.50* | 0.54* | 0.46* | 0.57* | |||||||
6. SD4 narcissism | 0.23* | 0.25* | 0.28* | 0.27* | 0.42* | ||||||
7. SD4 psychopathy | 0.49* | 0.36* | 0.48* | 0.50* | 0.32* | 0.37* | |||||
8. SD4 sadism | 0.66* | 0.52* | 0.61* | 0.68* | 0.51* | 0.31* | 0.57* | ||||
9. PSCD grandiose-manipulative | 0.37* | 0.48* | 0.39* | 0.46* | 0.46* | 0.64* | 0.32* | 0.33* | |||
10. PSCD callous-unemotional | 0.57* | 0.50* | 0.50* | 0.60* | 0.45* | 0.34* | 0.41* | 0.52* | 0.41* | ||
11. PSCD daring-impulsive | 0.42* | 0.38* | 0.50* | 0.48* | 0.34* | 0.44* | 0.54* | 0.47* | 0.42* | 0.37* | |
12. PSCD conduct-disorder | 0.52* | 0.47* | 0.57* | 0.58* | 0.30* | 0.32* | 0.58* | 0.54* | 0.37* | 0.44* | 0.53* |
- * p ≤ 0.001.
Males | Females | t (df) | p | Cohen's d | |
---|---|---|---|---|---|
M (SD) | M (SD) | ||||
Hurting people | 7.52 (5.77) | 4.69 (5.07) | −4.855 (380) | 0.001 | −0.53 |
Lying | 6.35 (4.57) | 4.06 (3.62) | −5.285 (380) | 0.001 | −0.58 |
Playing tricks | 3.27 (2.99) | 1.45 (2.27) | −6.571 (380) | 0.001 | −0.72 |
4 Discussion
During the last few years, research has focused attention on malevolent creativity, which encompasses dark aspects of creativity because of its malicious purposes that may cause harm to individuals and society (Batey et al. 2022). Despite this, there are few tools that can be used to measure this type of malevolent creativity (Kapoor and Khan 2016). Throughout this scenario, the Malevolent Creative Behavior Scale (MCBS) has been extensively utilized. In this paper, we adapted the measure to Brazilian Portuguese and tested its adequacy. Results indicated that the scale would be suitable for use in its Portuguese version.
Similar to the original version (Hao et al. 2016), the internal structure of the MCBS consisted of a three-factor model, which includes hurting others, lying, and playing tricks. All items presented good loading values and were kept in the Brazilian version. Additionally, Cronbach's alpha coefficients were found to be satisfactory, which were also higher than those reported by Hao et al. (2016), Tang et al. (2025), Ceballos and Watt (2023), and Zhao et al. (2022).
As part of this study, we evaluated the criterion-related validity of the Brazilian Portuguese version of the MCBS using the aversive personality traits. This choice was made in light of studies that revealed the dark side of creativity from the perspective of the Dark Triad, which links them together (Jia et al. 2020). In addition to the present study, some other studies have demonstrated that malevolent creativity is significantly associated with the Dark Triad (Gao, Cheng, et al. 2022; Kapoor and Khan 2016; Perchtold-Stefan et al. 2023). A relationship has also been found between conduct disorders and malevolent creativity. Previous findings indicate associations between malevolent creativity and domains related to interpersonal relationships, affective functioning, lifestyle, and antisocial behavior (Batey et al. 2022). This suggests that when an individual displays malevolent behavior, it can manifest itself as a disorder or as negative creative ideas, thereby justifying the positive correlation between the two constructs. As such, malevolent creativity can be associated with deviant behavior (Gáscon and Kaufman 2010).
One plausible explanation for these associations is the shared lack of moral restraints and empathy among individuals with higher levels of socially undesirable traits (Lee and Dow 2011; Mitchell and Reiter-Palmon 2022). For example, the interpersonal callousness typical of psychopathy (Gao, Qiao, et al. 2022) and the emotional detachment linked to callous-unemotional dispositions (Salekin and Hare 2016) may enable these individuals to generate creative yet harmful ideas without the usual moral constraints. In a similar vein, Machiavellianism and grandiose-manipulative tendencies, which prioritize the strategic exploitation of others (Paulhus et al. 2021; Pechorro et al. 2025), correspond with a tendency to fabricate lies and devise manipulative strategies. Overall, these findings suggest that malevolent creativity emerges when imaginative processes intermingle with personality traits that lack moral and empathic protections (Hao et al. 2016; Perchtold-Stefan, Rominger, and Fink 2023).
Based on our investigation of gender differences, significant differences were found among the groups. Males had higher means than females. Previous research has also shown that malevolent creativity differs by gender, with men being more negatively creative than women (Al-Mahdawi et al. 2022; Dumas and Strickland 2018; Harris and Reiter-Palmon 2015; Jia et al. 2020; Kapoor and Khan 2016; Kappor and Khan 2017; Lee and Dow 2011; Zhao et al. 2022). In accordance with the studies, men are more likely than women to develop harmful ideas.
Given that the results replicate global research on the subject, as well as those that use the instrument, including a significant relationship with aversive personality traits and gender differences, the findings motivate the continuation of the study. Considering that the process of evaluating the psychometric qualities of an instrument must be considered continuous, new studies must be conducted for use in the Brazilian population.
We recognized that part of the limitations of this study is because the study consists primarily of a convenience sample, with a higher percentage of women and without taking into consideration the educational level of the participants. Furthermore, the results associated with the Grandiose-Manipulative factor from the PSCD should be considered carefully, as its reliability was questionable or less than adequate. Despite the fact that the MCBS presented expected scores for the Brazilian population, further studies are recommended in which the sample may be diversified. Moreover, a cyber version (Liu et al. 2024) could also be investigated in future studies.
In addition to the limitations and recommendations already mentioned, we also suggest that future research explore the relationship between scores on the revised MCBS and other relevant factors, such as trait aggression, general creativity, and morality. While the present study focused on the relationship between malevolent creativity, the Dark Triad traits, and aversive personality characteristics, analyzing these additional aspects could contribute to a broader understanding of the underlying mechanisms of malevolent creative behavior. Including these new and important constructs would allow for an exploration of how different dimensions of personality and creativity are associated with malevolent behaviors and their implications in various social and individual contexts.
Ethics Statement
The complete project was developed according to the Declaration of Helsinki and its later amendments, and it was approved by an Ethical Board Committee in Brazil.
Conflicts of Interest
The authors declare no conflicts of interest.
Open Research
Data Availability Statement
Data is available upon reasonable request.