Development and initial validation of the Schadenfreude-by-concern inventory in a Japanese sample

Schadenfreude is a social emotion of passively rejoicing in the misfortune of others (Ben-Ze’ev, 2000). Since this emotion is a type of malice, public statements are avoided and rarely come forward. However, one has a propensity to take pleasure from the misery of others. In fact, there are numerous findings in psychology and cognitive neuroscience that recognize the existence of Schadenfreude (Brigham et al., 1997; Takahashi et al., 2009).

For the past 30 years, research in the field of social psychology has focused on identifying the conditions and situations that produce these emotions (e.g., van Dijk et al., 2006). Wang et al. (2019) reviewed empirical research and proposed a tripartite motivational model. This model suggests that people who rejoice in the misfortune of others are characterized as deviating from the norm, being envious, or belonging to a hostile group. In other words, Schadenfreude is more likely to be evoked by justice, envy, or social identity. Schadenfreude driven by justice is an emotional reaction of ridicule at the misfortune that befell the unjust, because they deserved it. This is called the justice-type Schadenfreude because it resembles a context in which one criticizes those who violate culturally shared norms. Schadenfreude driven by envy rejoices in the failures of others and reduces concerns about self-esteem. This is called the rivalry-type Schadenfreude, because it is based on rivalry in social comparison. Schadenfreude driven by social identity is an emotion that arises from social comparison as well as rivalry-type Schadenfreude. However, it is distinguished from rivalry-type Schadenfreude in that it requires intergroup conflict as a prerequisite for emotional arousal and is treated as an aggressive-type Schadenfreude. According to Wang et al. (2019), these three types of Schadenfreude can be classified in terms of two axes: the dimension in which the emotion arises (whether it arises on an individual basis or on the basis of group relations) and the direction in which the emotion goes (for oneself or for society). In other words, justice-type Schadenfreude occurs on an individual basis and is oriented toward society as a whole; rivalry-type Schadenfreude occurs on an individual basis and emphasizes self-oriented objectives; and aggressive-type Schadenfreude is based on group relations and arises for one’s own benefit. Wang et al. (2019) claim that since these differences in the nature of these Schadenfreude are recognized, it is necessary to treat them differently.

In measuring Schadenfreude, many previous studies have used the scene assumption method (Pietraszkiewicz, 2013; Smith et al., 1996). For example, Smith et al. (1996) facilitated social comparisons among participants in an experiment in which a fictitious person who was set to attend medical school, was presented in a successful situation (e.g., high grades or playing tennis with an attractive girlfriend). Then, in a later episode, Shill informs the participants that she has fallen into a misfortune (her admission was revoked) because of a misdeed (stealing amphetamines from the laboratory), and has them respond to the degree of Schadenfreude. Such a method provides information about the circumstances and conditions under which the emotion arises, but does not measure whether participants are more likely to feel Schadenfreude as a personality trait. Even if the situation encountered is the same, there may be individual differences in the degree of emotional response. In fact, since Spielberger (1966) argued that we should distinguish between state emotions and trait emotions, anxiety (Spielberger et al., 1970) and envy (Gold, 1996) have come to be considered not only as emotional responses to situations, but also in terms of the ease with which we experience these emotions. In the case of Schadenfreude, taking into account not only the state but also the trait aspect will be useful not only for identifying the factors that trigger the emotion, but also for predicting the motives and behaviors that follow it.

Scales that measure the trait aspects of Schadenfreude have been developed in recent years. The Chronic Schadenfreude Scale (CSS) by Krizan and Johar (2012) is a pioneer in this field. Although the CSS contains items that can measure components that Wang et al. (2019) refer to as rivalry-type Schadenfreude, it does not measure Schadenfreude derived from other aspects. In addition, since the CSS was composed of items that were created on a temporary basis as part of a narcissism study, no formal standardization procedures were followed. The Trait Schadenfreude Scale (TSS) developed by Crysel and Webster (2018) has identified a two-factor structure with reliability and validity: malicious Schadenfreude, which is likely to lead to more suffering in the person toward whom the Schadenfreude is directed, and benign Schadenfreude, which is used as humor in communication situations. However, most of the items in the TSS are ad hoc in terms of the characteristics and position of the person who suffers the misfortune and do not describe the preconditions that brings about the said misfortune. Therefore, it is not easy for people to concretely imagine Schadenfreude as they experience it in their daily lives.

As previously mentioned, the occurrence of Schadenfreude is influenced by the cognitive understanding of the reasons that lead to the misfortune and by the relationship with the person who suffers the misfortune. Therefore, when measuring the trait aspect of Schadenfreude, one cannot ignore the background factors that it evokes. However, no scale has been developed that can measure subtypes of Schadenfreude after specifying the context in which the emotion arises. Clearly describing the reason for the unhappiness and the relationship between the unhappy person and themselves in the item text allows respondents to imagine their experience more realistically. This would provide more meaningful data regarding the trait Schadenfreude and lead to further understanding of the phenomenon. Furthermore, we would be able to examine which people are more likely to experience schadenfreude in which context, and which behavior this emotion lead to. Given that typical antecedents for the elicitation of Schadenfreude have been identified (Smith et al., 2009; Wang et al., 2019), trait Schadenfreude refers to the tendency to experience the emotion in response to such typical situations. Therefore, in this study, we regard trait Schadenfreude as the extent to which individuals are prone to feeling Schadenfreude when placed in such situations, taking into account the typical antecedents of Schadenfreude.

The purpose of this study is to develop the Schadenfreude-by-Concern-Inventory (SCI) in a way that relies on the context of Wang et al. ’s (2019) model. In doing so, this study followed the following steps: First, to classify Schadenfreude into three types (justice, rivalry, and aggression), we prepared items with reference to the existing measures of justice, envy, and social identity. The factor structure, internal consistency, and construct validity were then examined. To examine the validity, we tested the following five hypotheses.

Previous studies have consistently reported that Schadenfreude is easily evoked in deserved misfortunes (Smith, 2013). Individuals who value the norms of their culture are likely to promote aggressive attitudes toward others who violate them. In other words, the more justice-oriented a person is, the more likely he or she is to rejoice in the misfortune caused by deviation from the norm. There are two different sub-concepts (Belief in Immanent Justice: BIJ and Belief in Ultimate Justice: BUJ) in just world beliefs that are believed to be involved in justice (Maes & Schmitt, 1999). BIJ is the tendency to attribute the cause of a negative event to a past wrongdoing committed by the person in question, with justice inherent in the resulting outcome (Hafer & Bègue, 2005). On the other hand, BUJ is the tendency to believe that the disadvantage suffered due to injustice will be compensated in the future and does not involve the tendency to attribute the cause of the damage (misfortune) to the victim (others who have been unhappy) or to blame the victim for their behavior (Murayama & Miura, 2015). Given the above, justice-type Schadenfreude is predicted to be positively correlated only with intrinsic just-world beliefs, since it is an emotion that rejoices in the misfortune of the unjust with the perception of deservingness (Hypothesis 1).

Previous studies have consistently reported a negative correlation between envy and self-esteem (Sawada & Fujii, 2016; Smith et al., 1999). For individuals with a stable self-image, others are not as important as the indicator of their self-worth, and envy is thought unlikely even if they witness others who are superior to them. Therefore, high self-esteem is predicted to contribute to the suppression of perceived ease of rivalry-type Schadenfreude (Hypothesis 2).

According to Lange and Crusius (2015), the envy subtypes are classified according to their differential effects. Two types are distinguished: malicious envy, which leads to motives aimed at undermining the superior opponent’s position and benign envy, which leads to motives aimed at improving one’s own position. Given these differences in the nature of envy, there should be a positive correlation between susceptibility to rivalry-type Schadenfreude, which is caused by a potential competitor pulling one’s leg, and susceptibility to malicious envy (Hypothesis 3).

We emphasize the superior aspects of the in-group to satisfy our desire for a positive social identity, which sometimes leads to discrimination against the out-group (Tajfel & Turner, 1979). In other words, the awareness of the existence of hostile external groups strengthens the unity of the internal group. Therefore, it is possible that a sense of belongingness to a group may suggest a positive correlation with the aggressive type (Hypothesis 4).

We also examine the relationship with the existing Schadenfreude scale; the SCI does not look at the benign aspect, which is attempted in the TSS. Thus, the positive correlation with the SCI subfactor, malicious Schadenfreude, is expected to be higher than that with benign Schadenfreude. However, feeling Schadenfreude is likely to be decoupled from a sense of malice, since the righteous tend to rejoice in the misfortune of those who violate the norm. Thus, malicious Schadenfreude may not be associated with justice-type Schadenfreude (Hypothesis 5).

The survey was conducted in December 2021 using an online survey service (Freeasy) provided by I-Bridge Inc. The study used the Directed Questions Scale (hereafter DQS; Maniaci & Rogge, 2014) to detect satisfaction trends in online surveys, adding two items to the survey content and excluding from the analysis the responses of those who violated the DQS even once. DQS is a method for detecting respondents who do not read the questions carefully. Specifically, a question is included in the Likert scale that instructs the respondent to select a specific response (e.g., “For this question, please select ‘1. Not at all applicable’”). It is determined whether the same response as the specified response has been selected. Previous research has shown that the responses of DQS violators disrupt the factor structure of psychological scales (Miura & Kobayashi, 2018). In this study, we used the DQS to detect respondents who did not read the questions carefully in order to improve the quality of the data. In this study, survey participants were recruited twice, and the DQS violation rate was 31% the first time and 14% the second time. The second survey was conducted for those who did not participate in the first survey. This was conducted because the number of participants in this survey did not reach the predetermined number.

Regarding the sample size for factor analysis, Kline (1993) argues that 4–10 participants per item are required. If we follow this standard, then for the SCI, which has 21 candidate items, we would need between 84 and 210 participants. In this study, we set the standard as obtaining 210 valid respondents.

Initially, 300 participants were recruited through the online survey service, 93 of them violated the DQS. An additional 50 participants were recruited, 7 of them violated the DQS. These procedures resulted in a total of 250 valid responses (175 males and 75 females). The mean age of the participants was 55.28 years (SD = 12.84). The 250 valid respondents were randomly selected from among the Japanese people living in Japan’s 47 prefectures. The age range of the valid respondents was 20–85 years old. More detailed demographic data is shown in Table 1.¹

Respondents who answered all the questions were awarded points (worth approximately 1–11 yen each) that were given within the relevant site via the points site registered by each respondent.

Item analysis was conducted assuming the theoretically predicted three dimensions. Mardia’s MVN test was then performed prior to exploratory factor analysis (EFA). The results showed that both multivariate skewness (chi.skew = 5042.69, p < .001) and multivariate kurtosis (z.kurtosis = 45.36, p < .001) were significant, indicating that the data under analysis deviated from a multivariate normal distribution. Therefore, the EFA for initial validation of the factor structure of the SCI was conducted using the maximum likelihood method with robust standard errors (MLR), geomin rotation. Cronbach’s α and McDonald’s ω coefficients for each factor of the SCI were calculated to assess internal consistency. To examine the construct validity of the SCI, correlation coefficients and partial correlation analysis were calculated between the SCI and the external measures (J-BQ, J-RSS, J-BeMaS, GGITS, and J-TSS). Since there was a significant positive correlation between the subscales of the SCI, partial correlation analysis was conducted. Therefore, when examining the correlation between the scales used as validity indicators and “justice-type Schadenfreude”, we calculated the partial correlation coefficient while controlling for “self-oriented-type Schadenfreude”. Similarly, when examining the correlation between other scales and “self-oriented-type Schadenfreude”, we calculated the partial correlation coefficient while controlling for “justice-type Schadenfreude”. When there is a correlation between the subscales of the SCI, when examining the correlation between each subscale and an external variable, there is a possibility that a pseudo-correlation affected by other subscales will be reflected. Finally, t-tests were conducted to examine gender differences in the SCI subscales.

The EFA was conducted by Mplus 8 (Muthén & Muthén, 2017). The computation of Cronbach’s α and McDonald’s ω is done by loading the psych package (Revelle, 2016) on the open-source statistical software R Version 4.2.2 (R Core Team, 2021). Loading the psych package (Revelle, 2016) on the open-source statistical software R Version 4.2.2 (R Core Team, 2021). Multivariate normality tests were performed by loading the MVN package (Korkmaz et al., 2014) on R version 4.2.2 (R Core Team, 2021). Other analyses were performed by using the statistical software HAD developed by Shimizu (2016).

Ethical considerations were explained in writing at the top of the online survey screen, and consent was obtained from all participants. The study was approved by the ethics committee of the first author’s affiliated university (approval number: 2021-25-02).

We conducted an item analysis assuming three dimensions. As shown in Table 2, alpha coefficients ranged from .84 (Aggressive) to .94 (Rivalry). Excluding any of the items did not contribute to an increase in the α coefficient of the subscale. The correlation between each item and the subscale ranged from .65 (Aggressive and Justice) to .88 (Rivalry).

A CFA was performed to see how well the theoretically assumed three-factor model fits the actual data. CFA was performed by loading the lavaan package (Rosseel, 2012) on R Version 3.6.2 (R Core Team, 2021), an opensource statistical software environment.

The analysis showed that the model fit was not satisfactory (χ² = 645.777, p = .000, CFI = .868, RMSEA = .104 (90% Confidence Intervals (CI) = [.095, .113]), AIC = 10,007.829). This suggested that the structure of the SCI might not be three factors, so the number of factors was explored by referring to the eigenvalues. The eigenvalues were, in order from the first factor, 11.07, 2.61, 0.91, 0.73, 0.59…, suggesting two factors for the Guttman and Scree plot criteria. Since two factors were proposed based on the results of the parallel analysis and the minimum mean partial correlation, the number of factors was specified as two in the EFA (MLR, Geomin rotation). 21 items of the EFA were conducted with two factors specified, and three items had factor loadings of less than 0.50 for any of the factors, indicating that these items had multiple loadings. Therefore, the same factor analysis was conducted after excluding these items. As a result, the first factor consisting of 12 items and the second factor consisting of 6 items were extracted. Specifically, the items that were initially expected to load on each of the rivalry- and aggressive-type Schadenfreude factors were comprehensively explained from the first factor. The second factor was then composed of items that were prepared assuming that they would be explained from the justice-type Schadenfreude factor. Since Wang et al. (2019) referred to rivalry-type and aggressive Schadenfreude as self-oriented emotions; we named the first factor corresponding to the self-oriented-type Schadenfreude. The second factor corresponded to the justice-type Schadenfreude proposed by Wang et al. (2019), and is therefore referred to as such in this study. The goodness of fit for this model was acceptable (χ² = 200.123, df = 118, p = .000, CFI = .966, TLI = .955, RMSEA = .053 (90% CI = [.040–.065]), SRMR = .027). Therefore, the most interpretable solution was the two-factor model, with 18 items (two factors) was selected as a candidate for SCI. The inter-factor correlations were significant (r = .49, p < .001). The results of the EFA, descriptive statistics, and excluded items are shown in Table 3.³

The Cronbach’s α and McDonald’s ω coefficients were high for the self-oriented-type Schadenfreude factor (α = .96, ω = .96), and justice-type Schadenfreude factor (α = .89, ω = .89).

The correlation coefficients and partial correlation coefficients between the subscales of the SCI and other variables are shown in Table 4. Incidentally, as previously mentioned, all of the external indicators used for SCI verification had acceptable internal consistency.

In the association between the justice-type Schadenfreude factor of SCI and Belief in Immanent Justice, a single correlation was not significant (r = .11, p = .072, 95% CI [−.01, .24]), but a positive correlation was significant in the partial correlation after self-oriented-type Schadenfreude control (r = .22, p < .001, 95% CI [.10, .34]). However, neither the correlation coefficient nor the partial correlation coefficient between the justice-type Schadenfreude and Belief in Ultimate Justice was significant. Thus, Hypothesis 1 (justice-type Schadenfreude and Belief in Immanent Justice will be positively correlated.) was supported.

From the factor analysis described above, rivalry-type and aggressive-type Schadenfreude, as claimed in the theoretical model of Wang et al. (2019), were integrated into self-oriented type Schadenfreude factor, so hypotheses 2–5 were examined for correlations between this and the other variables. In other words, based on the results of the EFA, we decided to replace the rivalry-type and aggressive-type Schadenfreude in Hypotheses 2–5 with self-oriented-type Schadenfreude.

First, the negative correlation between self-oriented type Schadenfreude of SCI and Self-esteem was significant, both before (r = −.27, p < .001, 95% CI [−.38, −.15]) and after (r = −.22, p < .001, 95% CI [−.34, −.10]) controlling for justice type Schadenfreude. Thus, Hypothesis 2 (self-oriented-type Schadenfreude and self-esteem will show a negative correlation.) was supported. Next, we examined the correlation between the self-centered-type Schadenfreude and benign and malicious envy. The results showed no significant correlation with benign envy (r = .06, p = .311, 95% CI [−.06, .18]) and a strong significant positive correlation with malicious envy (r = .76, p < .001, 95% CI [.70, .81]), as in Hypothesis 3 (There is probably a positive correlation between self-oriented-type Schadenfreude and malicious envy.). Incidentally, self-oriented-type Schadenfreude and malicious envy showed a strong and significant positive correlation (r = .76, p < .001, 95% CI [.70, .81]) even after controlling for justice-type Schadenfreude. On the other hand, focusing on the correlation and partial correlation coefficients between self-oriented-type Schadenfreude and General Group Identification Tendency, contrary to hypothesis 4 (The self-oriented-type Schadenfreude and the General Group Identification Tendency will be positively correlated.), the negative correlation was significant for the single correlation (r = −.21, p < .01, 95% CI [−.33, −.09]), and the partial correlation was not correlated (r = −.12, p = .057, 95% CI [−.24, .00]).

A significant positive correlation was found between the self-oriented-type Schadenfreude and benign and malicous Schadenfreude (benign Schadenfreude: r = .52, p < .001, 95% CI [.42, .60]; malicious Schadenfreude: r = .70, p < .001, 95%CI [.63, .76]). The positive correlation remained significant after controlling for justice-type Schadenfreude (benign Schadenfreude: r = .48, p < .001, 95% CI [.38, .57]; malicious Schadenfreude: r = .66, p < .001, 95% CI [.58, .72]). Regarding the correlation between the justice-type Schadenfreude and benign and malicious Schadenfreude, the correlation was significant before controlling for the self-oriented-type-type Schadenfreude (benign Schadenfreude: r = .22, p < .01, 95% CI [−.03, .22]; malicious schadenfreude: r = .33, p < .001, 95% CI [ .14, .37]). However, after controlling for self-oriented-type Schadenfreude, the correlation disappeared (benign Schadenfreude: r = −.08, p = .219, 95% CI [−.20, .05]; malicious Schadenfreude: r = −.07, p = .308, 95% CI [−.19, .06]). Thus, Hypothesis 5 (self-oriented-type schadenfreude is more strongly correlated with malicious schadenfreude than with benign schadenfreude.) was supported.⁴

Welch’ t test was conducted to examine gender differences in the SCI subscales. No significant differences were found between males (M = 2.61, SD = 0.77) and females (M = 2.57, SD = .0.80) for self-oriented-type Schadenfreude (t (135.50) = 0.38, p = .70, Cohen’s d = 0.05), and no significant difference was found between males (M = 3.49, SD = 0.79) and females (M = 3.46, SD = .0.77) for justice-type Schadenfreude (t (144.70) = 0.28, p = . 78, Cohen’s d = 0.04).

The incompatibility of the three-factor model itself means, in other words, that rivalry- and aggressive-type Schadenfreude are not distinct concepts to begin with. This issue is particularly important when focusing on the functions of emotions. As noted by Wang et al. (2019), the justice-type Schadenfreude has a different emotional direction than the other two. It functions for the benefit of society as a whole, whereas rivalry- and aggressive-type Schadenfreude are emotions brought about by social comparison and both function to restore self-esteem and self-exaltation. From an evolutionary perspective, emotion is a system that enhances individual fitness (Cosmides & Tooby, 2000). It has been noted that social status (Gilbert, 1990), which facilitates access to all resources, and cooperative behavior beyond kinship (Trivers, 1971) were important for human survival and reproductive success. At the same time as our minds have been designed to compete with rivals and seek social status, they have also evolved to cooperate with others. The two survival tasks (surviving social competition and maintaining social order) are important from the perspective of evolutionary adaptation, and when we perceive that a threatened survival task has been restored, we may experience Schadenfreude. If so, rivalry- and aggressive-type Schadenfreude would correspond to social status, and justice-type Schadenfreude to cooperative behavior, and from this perspective, the results of this study appear to be reasonable. However, the above interpretations are only speculative, and further studies are needed to determine whether social psychological theories and data can be explained by evolutionary psychology.

In terms of the limitations of the measurement method, the present results might reflect the difficulty of measuring whether emotions occur on an individual or group basis in a questionnaire. According to Wang et al. (2019), the difference between rivalry- and aggressive-type Schadenfreude lies in whether Schadenfreude is caused by comparison between individuals or assumes comparison between groups. In the present study, items that were prepared with the expectation that they would be explained by the aggressive-type schadenfreude factor were expressed, for example, “group,” “political party,” and “sports team,” and were qualitatively differentiated from items that questioned Schadenfreude through individual comparison (e.g., “jealous person” and “person with superior characteristics”). However, even if these item expressions were written in the question text, it is not certain that they would have imagined the psychological experiences that occur in everyday life. Research on the distinction between in- and out-groups has been conducted through experimental studies. For instance, Tajfel et al. (1971), who conducted a pioneering study on differences in behavior toward in- and out-group members, had experimental participants perform a test for group division (e.g., judging preferences between two types of abstract images on a screen) to increase their sense of belonging to an in-group. When the sense of belonging to an in-group increases, its superior aspects are emphasized (in-group bias), which tends to discriminate against the out-group (Tajfel & Turner, 1979). Such intergroup dynamics can only be embodied by experimentally categorizing groups. Studies showing that people are more likely to feel Schadenfreude in response to out-group misfortunes have also been conducted experimentally in ways that specifically mirror the out-group (e.g., Cikara et al., 2011). If the reason for the integration of rival- and aggressive-type Schadenfreude into one dimension is the limitations of the measurement method, future research should return to the discussion of whether questionnaires can measure intergroup dynamics and accumulate methodological knowledge.

Next, we discuss the relationship between SCI and external measures. Consistent with theoretical expectations, justice-type Schadenfreude was positively correlated with Belief in Immanent Justice and not with Belief in Ultimate Justice. People tend to experience Schadenfreude when they judge that the misfortune that befell another person was based on that person’s unethical behavior (Smith, 2013). According to Murayama and Miura (2015), the higher a person’s level of internalized fairness beliefs, the stronger their tendency to punish perpetrators. In light of this, the tendency to experience justice-type Schadenfreude could lead to aggressive behavior. Moreover, the act of rejoicing a person’s misfortune is less likely to be negatively evaluated by other group members when it is in accordance with culturally shared standards (Watts, 2008). In other words, the tendency to experience justice-type Schadenfreude is likely to be tolerated by the group. Therefore, it is assumed that the tendency to experience justice-type Schadenfreude spreads easily within the group. Clarifying these points would be beneficial for further strengthening the validity of justice-type Schadenfreude. Future research is expected to incorporate not only cross-sectional survey research using interindividual correlations but also research methods that consider the influence of group dynamics.

Interestingly, Belief in Immanent Justice is negatively correlated with self-oriented-type Schadenfreude factor with and without control of justice-type Schadenfreude factor. This result might be due to the fact that it was evaluated as evil and the Belief in Immanent Justice as pure-hearted. Self-oriented-type Schadenfreude and Belief in Immanent Justice differ in that the former is an emotion that rejoices in one’s own benefits at the expense of others, while the latter refers to the tendency to believe that one’s benefits at the expense of others will eventually be made up. Those who want the world to be just and will not tolerate evil that threatens it may be less likely to agree with self-oriented-type Schadenfreude’s items, which tend to give an impression of selfishness and self-centeredness.

Self-oriented-type Schadenfreude, as expected, correlated negatively with Self-esteem and positively with malicious envy. According to Takahashi et al. (2009), who used fMRI to elucidate the neural basis of envy, the brain pain system is activated when humans experience envy. Self-oriented-type Schadenfreude may function to compensate for perceived deficiencies and distress in comparison to superior others. A meta-analysis of the association between envy subtypes and Schadenfreude also suggests that Schadenfreude is strongly positively correlated with malicious envy (Lange et al., 2018). Since self-oriented-type Schadenfreude is a pleasant feeling aroused by misfortune that befalls a person who is the object of envy, the positive correlation between malignant envy and self-oriented-type Schadenfreude is also consistent with theoretical predictions. And self-oriented-type Schadenfreude can be rephrased as a personality prone to deriving pleasure from the misfortune of a person who is the object of envy. Therefore, the positive correlation between malicious envy and the self-oriented-type Schadenfreude factor aligns with theoretical predictions. Additionally, since self-oriented-type Schadenfreude is based on envy, the negative correlation with self-esteem, which is known to be inversely correlated with envy, also makes sense.

Hypothesis 4 was that Schadenfreude, due to social identity, would be positively correlated with General Group Identification Tendency. The lack of a positive correlation between self-oriented-type Schadenfreude and group identification tendency did not support Hypothesis 4. As noted above, factor analysis did not discriminate between rival- and aggressive-type-Schadenfreude, so it was not possible to distinguish whether Schadenfreude occurred between individuals or based on group relationships. In particular, if the reason for this is to be sought in the limitations of the measurement method, it is possible that it was difficult to measure the psychological dynamics of the group dimension that affected the results. Therefore, items designed to measure the characteristics that occur between groups may not accurately measure the pre-assumed meanings. Hypothesis 4 may have been incorrectly conceived, even if they did not cancel out. For example, one study measured the emotional reactions of enthusiastic baseball fans to situations in which a team other than the one they cheer to make a mistake (Cikara et al., 2011). The experimental collaborators, baseball fans, showed a strong joy response when they witnessed the failure of the rival team (e.g., the Red Sox in the case of Yankees fans), indicating the presence of social identity-derived Schadenfreude in the intergroup struggle. At the same time, however, it became clear that there was little joy in witnessing the failure of a non-rival team. This can be seen as a corollary that what played an important role in arousing Schadenfreude was not a sense of belonging to the team one was rooting for, but hostility toward the rival team. The General Group Identification Tendency (Matsuki & Shimotsukasa, 2020) used in this study measures the tendency to identify with the in-group in general and does not provide information on the tendency to have hostile attitudes toward the out-group. Moreover, this scale is composed of items that easily evoke positive impressions. Considering this point, the finding in this study that the tendency for group identity is negatively correlated with the tendency to experience self-oriented-type Schadenfreude makes sense. In the future, we must examine the nature of Schadenfreude brought about by social identity, not only in terms of belonging to an in-group, but also in terms of hostility toward an out-group and stereotypical attitudes.

The fact that the self-oriented-type Schadenfreude factor of the SCI correlated positively with the J-TSS, an existing measure of Schadenfreude traits, provides support for the validity of the SCI. Of particular importance is the high positive correlation with malicious Schadenfreude, which is supposed to follow the definition of Schadenfreude in previous studies. Kato and Fujimori (2021) have shown that individuals with a higher tendency to feel malicious Schadenfreude are less likely to exhibit empathic concern for others. This suggests that a personality, which is prone to experiencing self-oriented-type Schadenfreude, may be associated with an attitude that disregards the interests and rights of others. However, in this study, we were unable to examine it as a possible factor in the interpersonal behavior index. Future research should incorporate, for example, other-derogatory behavior (e.g., spreading bad rumors) as a predictor variable in the analysis to determine if it is a trait that leads to behaviors that inhibit the rights of others.

It is interesting to note that the partial correlation between the tendency to experience justice-type Schadenfreude and malicious Schadenfreude is not significant. According to Boehm (2012), ridiculing deviant individuals not only socially excludes them but also prevents people with similar traits from engaging in the same deviant behavior. It may be because of these effects that the tendency to experience justice-type Schadenfreude is less likely to be perceived as outright malice. In a study dealing with Schadenfreude caused by feelings of envy (Sawada, 2008), it was revealed that guilt contributes to the arousal or suppression of the expression of Schadenfreude in women. However, even a guilt-prone person may not suppress the arousal or expression of Schadenfreude if it originates from justice rather than envy. In the future, it will be necessary to examine whether individuals who are less likely to experience self-oriented-type Schadenfreude and justice-type Schadenfreude are more prone to feeling guilt. If guilt only suppresses the tendency to experience justice-type Schadenfreude, it could also serve as evidence for the validity of the SCI.

The present results showed no gender differences in the SCI subscale. However, focusing on the actual process of generating the status Schadenfreude, it is possible that the context of emotional arousal may differ between men and women. Both men and women experience Schadenfreude more strongly when the person who has suffered the misfortune is of the same gender than when the person is of the opposite gender, but women are more likely to experience Schadenfreude in response to a decrease in physical attractiveness of the same gender and men in response to a decrease in social status of the same gender (van Dijk et al., 2015). Future research would also be beneficial in examining the detailed mechanisms by which an individual’s inherent tendency to experience Schadenfreude predicts how state Schadenfreude, which varies depending on the context, is evoked. In this study, as a result of random sampling, the proportion of male participants was higher than that of female participants. Since Welch’s test showed no significant differences between genders on both factors of the SCI, and the effect size was extremely low, it is unlikely that the imbalance in the number of participants had a significant impact on the results of this study. However, when considering factors such as age, social class, and place of residence, it is possible that differences in response tendencies for each factor of the SCI may emerge (i.e., young men in competitive environments may score higher on self-oriented Schadenfreude compared to others). Therefore, future research on the SCI should ideally consider balancing the number of respondents across social class, place of residence, age, and gender when conducting the survey.

This study has several limitations. SCI is a scale that measures the susceptibility to Schadenfreude as a personality trait, but in this study, we did not examine its retest reliability. Therefore, it cannot be said with certainty from the data in this study that the trait Schadenfreude that stabilize within individuals can be measured. Therefore, examining the reliability of the SCI from the perspective of stability will be an issue for the future. Another important issue would be to verify the measurement invariance of SCI. In this study, we examined the structure, reliability, and validity of the SCI based on data from Japanese people of a wide range of ages (from their 20s to their 80s). However, the ease of experiencing Schadenfreude may differ depending on age and place of residence. For example, it is possible that young people living in competitive urban centers would score higher on the self-oriented-type Schadenfreude. Further study is needed in this regard, such as obtaining data from different samples and conducting a simultaneous multigroup analysis using such data.

Moreover, when treating Schadenfreude as a trait variable, careful discussion is necessary regarding its differences from low agreeableness. In particular, another scale measuring trait Schadenfreude (TSS) has revealed a weak to moderate negative correlation between Schadenfreude and agreeableness (Crysel & Webster, 2018; Kato & Fujimori, 2021). Based on this, trait Schadenfreude appears to reflect a lack of agreeableness. However, as mentioned earlier, trait Schadenfreude is ultimately the tendency to experience an emotion in response to specific situations and is not necessarily synonymous with low agreeableness. For example, individuals who tend to value harmony with others may be more likely to feel joy in situations where norm violators receive punishment (i.e., they may score higher on justice-oriented Schadenfreude). In this way, the role of agreeableness may vary depending on the context in which Schadenfreude arises. However, since this study did not quantify agreeableness, further examination of this point will be necessary in the future.

Despite these limitations, this research is significant in that it has provided a measurement tool that can be used in future studies of Schadenfreude. In this study, we attempted to evaluate Schadenfreude not from the perspective of a state-like emotional reaction, but from the viewpoint of a personality trait. This approach is expected to not only clarify the conditions under which Schadenfreude is more likely to be triggered, as examined in previous studies, but also enable the prediction of motivations and behaviors associated with this emotion. In existing scales for measuring trait Schadenfreude, Schadenfreude is distinguished into benign Schadenfreude, where harm to the social relationship with the target is unlikely, and malicious Schadenfreude, which is more likely to cause harm (Crysel & Webster, 2018). The SCI developed in this study measures the susceptibility to trait Schadenfreude, taking into account the typical triggers or points of interest that precede the emergence of Schadenfreude. Therefore, it can be considered unique in that it allows for the quantification of susceptibility to Schadenfreude in relation to specific events and situational contexts. Schadenfreude, if not properly managed, can lead to harmful behaviors that unjustly degrade others (e.g., online defamation, malicious gossip). Measuring Schadenfreude while considering the differences in the contexts preceding the emotional response is valuable for uncovering how these emotions are linked to specific behaviors. Understanding these mechanisms is expected to provide important insights for clinical practices aimed at reducing problem behaviors potentially related to Schadenfreude.