Abstract
This study examined the moderating role of four humor styles, namely, self-enhancing, affiliative, self-defeating, and aggressive humor, in the relationship between perceived stress and physical health. The sample comprised 954 undergraduate students of both genders between the ages of 18 and 43 years (M = 20 years) from various ethnic backgrounds (68.2% Black African, 19.1% White, 7.3% Colored, and 5.3% Indian). Three self-report measures were administered to assess humor styles, perceived stress, and physical health. Results showed that the self-defeating humor style moderated the relationship between stress and physical health with higher use of this humor style associated with an increase in reported physical health symptoms. The results highlight the role of a self-directed, detrimental humor style in the deterioration of physical health during the perceived experience of stress, thus indicating that one’s habitual way of using humor could influence whether physical health outcomes are positive or negative.
Stressful experiences, especially those of a chronic nature, may contribute toward a variety of adverse health outcomes (Martin, 2002; Schat, Kelloway, & Desmarais, 2005). In this regard, some studies have found that humor could be a coping mechanism or moderator that lessens the impact of frequent stressors (Lefcourt, 2002; Martin, 2003; Svebak, 2010). These results were, however, contradictory, with some studies finding little or no beneficial outcomes (Gremigni, 2012). An emphasis on humor as a unidimensional construct has been proposed as playing a role in the lack of stable findings. Confirming the multidimensional nature of humor, Kirsh and Kuiper (2003) found empirical support for two broad sense of humor dimensions which indicate that humor can have both beneficial and detrimental aspects. From this viewpoint, it is possible that buffering of stress and concomitant enhancements in one’s health may not only occur as a result of constructive humor usage but also from the avoidance of harmful humor use (Cann, Stilwell, & Taku, 2010; Gremigni, 2012; Martin, Puhlik-Doris, Larsen, Gray, & Weir, 2003).
With regard to humor use, Martin et al. (2003) proposed four humor styles, two of which are believed to lead to favorable outcomes and two of which are thought to result in negative outcomes. Affiliative humor refers to humor which is affirming to oneself and others in a tolerant and nonaggressive manner through the use of jokes, witticisms, stories, and laughter. This style of humor functions on an interpersonal level and is often used to foster social cohesion. Aggressive humor, on the contrary, refers to humor that is impulsive and derisive or disparaging toward others, such as sarcasm, teasing, and ridicule. These two humor styles are considered to be other-oriented.
The self-enhancing humor style is used to aid the individual in maintaining an optimistic outlook on life when presented with stressful or adverse circumstances. As a consequence, this style of humor is often regarded as a form of coping humor. The self-defeating humor style is used as a form of self-disparagement to gain others’ favor or approval by allowing oneself to be the focus of another’s jokes, as well as sharing and contributing to this amusement at one’s own expense. These two humor styles are considered to be self-oriented (Martin et al., 2003).
Past studies have examined the moderating role of sense of humor in the relationship between stress and physical health (e.g., Abel, 1998; Porterfield, 1987), but very little is known regarding the moderating role of the four humor styles in the stress–health relationship. Given the paucity of research focusing on the moderating role of the humor styles in the stress–health relationship and the inconsistency of findings thus far, the aim of the current study was to utilize a parsimonious model to determine whether the four humor styles moderate the effects of perceived stress on physical health.
Theoretical Background
Research regarding the beneficial role of humor in coping with stress gained impetus when Martin and Lefcourt (1983) reported three studies indicating that sense of humor could buffer the harmful effects of stressful life events on mood. Porterfield (1987), however, replicated their research and expanded the scope to include physical health but found no evidence for the stress-buffering effects of humor. In contrast, Abel (2002) reported that a high sense of humor was related to less perceived stress and concluded that a high sense of humor led to beneficial stress coping strategies, such as finding something amusing in stressful circumstances. This, in turn, could lower negative emotional reactions or it could provide an incentive to change a stressful situation. In further studies, evidence for the health and stress-buffering benefits of humor remained equivocal (see Martin, 2004, for a review).
Evidence for the health benefits related to a good sense of humor, regardless of stress factors, is also lacking consistency. For example, a prevalence study in Norway which focused on common bodily symptoms, such as nausea and blood pressure, subjective health satisfaction, and sense of humor found no significant relationships between humor and health indicators among 65,000 participants, although a very weak positive correlation between sense of humor and health satisfaction emerged (Svebak, Martin, & Holmen, 2004).
In contrast, Peterson, Park, and Seligman (2006) found in a web-based study of 2,087 adult volunteers between the ages of 35 and 44 that, in terms of recovering from a serious physical illness, such as autoimmune disease, infectious disease, cardiovascular disease, cancer, and debilitating pain, the use of humor predicted fewer detrimental outcomes in terms of recovery from physical illness. Similarly, in a recent 15-year follow-up study among 53,000 participants in Norway regarding sense of humor and mortality, it was found that humor assisted with recovery from cardiovascular diseases in women and infection-related diseases in men (Romundstad, Svebak, Holen, & Holmen, 2016).
In terms of the role of the four humor styles in physical health, evidence of humor having an indirect effect has been found. For example, a study examining the extent to which the humor styles and negative affect predict physical health found levels of negative affect to be associated with more physical symptoms, but none of the humor styles emerged as significant predictors when negative affect was controlled for. Self-enhancing humor did, however, emerge as a significant predictor of adaptive coping strategies, such as perspective shifting, whereas aggressive humor predicted denial and less willingness to shift perspective (Kuiper & Harris, 2009).
Cann et al. (2010) also found that humor styles could play an indirect role in physical illness. In their study, longitudinal data on humor styles, positive personality characteristics, perceived stress, and health outcomes were used to test a theoretical mediation model. Results supported the model and indicated that positive personality characteristics, such as hope and optimism, mediated the relationship between self-enhancing and self-defeating humor and perceived stress which, in turn, predicted beneficial or detrimental health outcomes, respectively.
Cann et al. (2010) concluded that humor styles which focus on the self showed greater potential in terms of influencing the relationship between stress and health than do the other-oriented humor styles. The authors hypothesized that the role of these humor styles may lie in promoting a certain style of thinking and behaving which encourages a more positive personality. Such a personality style could lead to a more optimistic worldview and a tendency to perceive one’s life as less stressful which could bring about better physical and psychological health. Cann et al. (2010) concluded that “other directed humor styles are of little importance in understanding the relationship between sense of humor and health outcomes” (p. 227).
Although the study by Cann et al. (2010) did not support other-directed humor styles as significant to the stress–health relationship, Martin (2002) postulated that the use of affiliative humor, an other-directed style, may have indirect stress-buffering and health-enhancing effects as a result of social facilitation and support. Some evidence for this was found by Kuiper and McHale (2009), who studied the role of humor styles as mediators in the relationship between self-evaluative standards and psychological well-being. It was found that higher positive self-evaluative standards led to more use of affiliative humor which resulted in higher social self-esteem and lower levels of depression. On the contrary, higher negative self-evaluative standards led to increased use of self-defeating humor which led to lower social self-esteem and higher levels of depression.
Kuiper and McHale (2009) concluded that more use of affiliative humor could assist in the development and maintenance of social support networks that foster and improve well-being. Thus, it appears as if the affiliative humor style serves more of a beneficial and adaptive function for the purpose of social cohesion and self-esteem than coping directly with stress. High use of self-defeating humor may also play an indirect role by leading to the development of maladaptive social support networks which could result in lower psychological well-being.
Since Martin’s (2004) review, it appears as if only one study has focused on humor styles as moderators in the stress–health relationship. Evaluating the stress-buffering role of humor styles in physical health, a study by Freeman and Ventis (2010) examined the moderating role of the humor styles in the relationship between stress and health, including hassles, among 265 retirees. For affiliative humor, a beneficial moderating effect was found regarding general health but only when hassles intensity was high. No significant moderating effects were found for self-enhancing humor in terms of physical health.
In terms of the moderating role of self-defeating humor regarding pain, the maladaptive effect of this humor style only emerged when daily hassles or stress were low. At high levels of daily hassles or stress, a high self-defeating score was related to less pain being reported. Similarly, aggressive humor was found to play an adaptive role when levels of hassles intensity or stress were high (Freeman & Ventis, 2010).
When gender was added as a moderator, significant results were again only found for the two maladaptive humor styles; however, the above-mentioned results were only found for men. For women, high levels of daily hassles or stress led to detrimental outcomes when self-defeating or aggressive humor use was high. Freeman and Ventis (2010) concluded that the detrimental effects of the aggressive and self-defeating humor styles might only emerge in certain situations and depending on gender. The authors called for more varied samples and different contexts to be included in future studies.
It is clear that the role of the individual humor styles in the stress–health relationship is still unclear, with some studies (e.g., Cann et al., 2010; Kuiper & Harris, 2009) indicating that other variables, such as personality characteristics, could mediate the effect of the humor styles in this relationship. The findings of Freeman and Ventis (2010) show that gender and contextual factors could play a role in terms of humor, stress, and health outcomes among elderly people.
The current study evaluated a parsimonious moderation model and stated the following broad hypotheses:
Method
Participants
A convenience sample, consisting of 954 introductory psychology students, was included in the study. Participants were Asian (1%), Colored (7.3%), Indian (4.3%), Black (68.2%), and White (19.1%). 1 The age of the participants ranged between 18 and 43 years (M = 20, SD = 2.478), and there were 702 females and 252 males.
Measures
To gather demographic information, a biographical questionnaire was used in which participants were required to indicate their age, gender, and cultural affiliation. Three self-report measures were used to assess individual humor styles, perceived stress, and physical health.
The Humor Styles Questionnaire (HSQ)
The HSQ (Martin et al., 2003) measures individual differences in styles of humor. It consists of four scales of eight items each, detailing different humor styles, two of which are other-directed (Affiliative and Aggressive Humor) and two of which are self-directed (Self-Enhancing and Self-Defeating Humor). Responses are measured on a scale ranging from 1 (totally disagree) to 7 (totally agree). Affiliative humor is assessed by questions such as, “I laugh and joke a lot with my closest friends,” while a sample item of aggressive humor is, “If I don’t like someone, I often use humor or teasing to put them down.” Self-enhancing humor is represented by questions such as, “Even when I’m by myself, I’m often amused by the absurdities of life,” while self-defeating humor is illustrated by questions such as, “I often go overboard in putting myself down when I am making jokes or trying to be funny.”
The HSQ has been shown to have satisfactory internal consistency on all four scales, with Cronbach’s alphas of .77 for the Aggressive Humor scale, .80 for the Affiliative Humor scale, .80 for the Self-Defeating Humor scale, and .81 for the Self-Enhancing Humor scale. In addition, the HSQ displayed satisfactory test–retest reliability coefficients of .81 to .85 on all four scales (Bilge & Saltuk, 2007; Hiranandani & Yue, 2014; Martin et al., 2003). Low intercorrelations between the four scales indicate distinction among these different dimensions (Martin et al., 2003).
The Perceived Stress Scale (PSS)
The PSS (Cohen, Kamarck, & Mermelstein, 1983) is a self-report measure of individual appraisals of stress. PSS items were designed to measure the extent to which respondents recognized their lives to be “unpredictable, uncontrollable, and overloading” (Cohen et al., 1983, p. 387). According to Cohen et al. (1983), these three aspects have been consistently found to be fundamental contributors to the experience of stress. The PSS consists of 10 items rated on a 5-point scale ranging from never (0) to very often (4). An example of how these items are presented is, “In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?” The PSS has been shown to have adequate internal consistency. In three samples of college students, Cronbach’s alphas ranging between .84 and .86 were established (Cohen et al., 1983). Test–retest reliability is indicated at .85 for this scale. In terms of predictive validity, the PSS was found to be suitable for the prediction of life events, depressive symptomology, and social anxiety (Cohen et al., 1983)
The Physical Health Questionnaire (PHQ)
The PHQ (Schat et al., 2005) is used to assess four specific somatic symptoms, namely, sleep disturbances, headaches, gastrointestinal problems, and respiratory illness. The PHQ consists of 14 items pertaining to the frequency with which respondents experience the aforementioned health-related issues. Items 1 to 11 are rated on a 7-point scale relating to the frequency of symptoms, namely, from 1 (not at all) to 7 (all the time). These items can be illustrated by questions such as, “How often did you get a headache when you were frustrated because things were not going the way they should have or when you were annoyed at someone?” Items 12 and 13 such as “How many times have you had minor colds (that made you feel uncomfortable but didn’t keep you sick in bed or make you miss work)?” are rated in terms of how many times the respondent has experienced specific ailments (0 to 7+ times). Item 14 is rated according to how long a bad cold or flu has lasted (1 to 7+ days). Higher scores on the PHQ indicate more physical health problems.
The reliability and validity of the PHQ were established in two separate samples. Internal consistency analyses of the four PHQ subscales revealed the following Cronbach’s alpha values for Studies 1 and 2, respectively: .84 and .86 for the Gastrointestinal Problems subscale, .90 in both samples for the Headaches subscale, .81 in both samples for the Sleep Disturbances subscale, and .70 and .77 for the Respiratory Illness subscale (Schat et al., 2005).
Procedure
During an introductory psychology lecture, students were advised about the study and its purpose and invited to participate. An incentive was offered in the form of additional course credit. Students were provided with the option to complete an assignment to receive the course credit, if they did not wish to participate in the study. Participants were required to access and complete the questionnaires on the university’s online student portal which is only accessible to registered students. Once logged into the portal, participants were provided with more information regarding the purpose of the study and the process of completing the questionnaire.
Data Analysis
Zero-order correlations were calculated first to explore the data and check for multicollinearity. A moderated multiple regression analysis was then performed to determine whether the relationship between perceived stress and physical health is moderated by the four humor styles. Where moderation was found for a specific humor style, a further moderated multiple regression analysis was performed to investigate the moderating role of the humor style in isolation from the other styles.
Ethical Considerations
Participants were briefed about the nature of the study and assured that their responses would be confidential. They were informed that the study was voluntary and that they could withdraw at any time without any negative consequences. Although participants had to log in to the online system to participate, their responses were anonymous. The study received institutional approval.
Results
Table 1 shows the means, standard deviations, and Cronbach’s alphas of the measures, as well as zero-order correlations among the variables of interest. Significant negative correlations were found between affiliative humor style and physical health (r = −.083, p < .05), as well as between self-enhancing humor style and perceived stress (r = −.164, p < .01). The self-defeating humor style showed significant positive correlations with both physical health (r = .142, p < .01) and perceived stress (r = .147, p < .01). Physical health showed a significant positive correlation with perceived stress (r = .440, p < .01).
Means, Standard Deviations, Alphas of the Measures, and Zero-Order Correlations Between Humor Styles, Physical Health, and Perceived Stress.
Correlation is significant at the .05 level. ** Correlation is significant at the .01 level.
A hierarchical moderated multiple regression analysis was performed with physical health serving as the dependent variable, perceived stress as the independent variable, and all four humor styles serving as moderator variables. The results are summarized in Table 2.
Perceived Stress, Physical Health, and the Moderating Effect of the Four Humor Styles.
Note. f2 = 0.30.
p < .05. **p < .01.
With perceived stress as the only predictor, R2 = .194, F(1, 960) = 230.878, p < .001. The four humor styles explained an additional 1.3% (
A hierarchical moderated multiple regression analysis was subsequently performed with physical health serving as the dependent variable, perceived stress as the independent variable, and self-defeating humor style as the only moderator variable. The results are summarized in Table 3.
Perceived Stress, Physical Health, and the Moderating Effect of the Self-Defeating Humor Style.
Note. f2 = 0.26.
p < .05. **p < .01.
With perceived stress as the only predictor, R2 = .194, F(1, 960) = 230.878, p < .001. Self-defeating humor style explained an additional 0.6% (
The moderating effect of self-defeating humor style on the relationship between perceived stress and physical health.
Figure 1 displays three statistically significant interaction lines, indicating that the relationship between perceived stress and physical health symptoms is stronger at higher levels of self-defeating humor. More specifically, at low levels of perceived stress, people who score high on the self-defeating humor style display fewer health problems compared with people who are considered to be low on self-defeating humor. However, as perceived stress increases, a reversal in outcomes occurs and greater use of the self-defeating humor style becomes associated with more physical health problems. A possible reason for this could be that, at low levels of perceived stress, people who employ greater use of self-defeating humor could be more realistic about their health. However, as stress increases, self-defeating humor may become more detrimental to the individual, owing to its negative nature. As perceived stress levels increase, a lower than average self-defeating humor style is associated with a lower increase in physical health symptoms. It seems, then, that an above average tendency toward using the self-defeating humor style could be beneficial for one’s health at low levels of perceived stress.
Discussion
A negative correlation between the affiliative style of humor and physical health symptoms was found. Thus, greater use of affiliative humor was related to fewer reports of physical health symptoms. This result is in keeping with previous findings, in that affiliative humor might play a role in the production of health-enhancing effects, probably as a result of social facilitation and social support (Cann & Etzel, 2008; Martin, 2007). Kuiper and McHale (2009) found, for example, that more use of affiliative humor could lead to the development and maintenance of social support networks that foster and improve well-being.
Of particular interest in this study was that no significant correlation was found between affiliative humor and perceived stress. This is in keeping with findings by Martin et al. (2003), who found that the affiliative humor style serves mostly a beneficial and adaptive function for the purpose of social cohesion and self-esteem rather than coping directly with stress. Therefore, it is possible that, in the present study, no significant correlation was found between affiliative humor and perceived stress as university students may use affiliative humor for the purposes of fostering new friendships and social acceptance rather than to cope with stress. Rezan (2007), for example, studied the relationship between the four humor styles and loneliness among university students. Results indicated a strong negative correlation between loneliness and affiliative humor. Overall, low use of affiliative humor was the best predictor of loneliness in this population. Similar findings that support the affiliative humor style as serving a predominantly social function, especially in novel situations, can be found in research by Hampes (2005) and Hiranandani and Yue (2014).
Self-enhancing humor style showed a negative correlation with perceived stress. Martin et al. (2003) found self-enhancing humor to be a beneficial style of humor, which assists the individual in maintaining an optimistic outlook during stressful times. More specifically, the relationship between self-enhancing humor and perceived stress in the current study may be due to the self-enhancing humor style involving perspective taking (Martin, 2007; Martin et al., 2003). Therefore, owing to the fact that the experience of stress relies strongly on perception (Lazarus & Folkman, 1984), this style of humor may assist in appraising stressful events as less threatening due to a generally positive outlook on these situations.
It is surprising that self-enhancing humor did not show a significant, negative correlation with physical health symptoms in the current study. As self-enhancing humor serves an adaptive function in terms of boosting the morale of the individual (Martin, 2007), it might be expected to be associated with decreased reports of physical health symptoms by buffering the individual against negative affectivity. However, while a review of the literature shows self-enhancing humor to be positively correlated with psychological well-being (e.g., Kuiper, Grimshaw, Leite, & Kirsh, 2004; Martin et al., 2003), there appears to be little evidence from past studies (e.g., Kuiper & Harris, 2009) to suggest that a relationship between self-enhancing humor and physical health exists. Unlike stress, physical health can be influenced by both psychological and physiological mechanisms, and therefore might involve a more complex form of cognitive manipulation concerning individuals’ perception of events.
In this study, there were positive correlations between the self-defeating humor style and both perceived stress and physical health symptoms. This finding is supported by results reported by Cann et al. (2010) which showed that self-directed humor that lowers a person’s self-worth was positively related to the level of stress experienced and, as a consequence, is negatively related to self-reported efficacious physical health. This finding is especially noteworthy in that it highlights that while the use of beneficial humor may allow for better coping with stressors and increased well-being, so too can harmful styles of humor precipitate perceived stress and result in poorer health outcomes. A possible reason for this could be that self-defeating humor can function as an avoidance strategy (Martin et al., 2003) and stress might therefore not be dealt with constructively, allowing it to increase and become more pervasive, resulting in damage to psychological and physiological systems.
Humor may provide a means of dealing with stress in an effective manner which, in turn, may enhance physical well-being (Martin, 2002, 2009). However, there is a possibility that certain styles of humor may be more stress-buffering and health-enhancing than others (Martin, 2002). Results from the current study showed that only self-defeating humor acted as a moderator between perceived stress and physical health. In other words, perceptions of stress and the incidence of negative physical health symptoms tend to increase if the self-defeating style of humor is used.
The self-defeating humor style is characterized by self-disparagement so as to gain social acceptance. Pervasive use of this style indicates emotional neediness, low self-esteem, and avoidance techniques (Martin et al., 2003). Thus, the negative attributes of self-defeating humor could provide an explanation as to why increased use of this style of humor is likely to result in detrimental outcomes in terms of perceived stress and physical health. If people actively avoid dealing with stress which is causing them anxiety, there is a possibility that the stress experience may be magnified, leading to additional problems, such as poor health. In addition, self-defeating humor is linked to neuroticism, anxiety, and depressed mood. Failing to deal with problems in a constructive manner may serve to increase the effects of these negative characteristics. More specifically, neuroticism is linked to greater experiences of stress and subsequent detrimental health behaviors (Mroczek, Spiro, & Turiano, 2009). Thus, greater use of self-defeating humor may amplify neurotic behaviors, leading to greater experiences of stress and poorer physical health. In support of this, Cann and Etzel (2008) reported that a sense of humor, which makes use of negative, self-directed forms of humor, acts to suppress the development of positive personality traits and increase the negative appraisal of stressful situations.
Self-defeating humor is used as a means to ingratiate oneself with others (Martin et al., 2003). Figure 1 shows that, at low levels of perceived stress, this ingratiation and subsequent feeling of belonging could act as a buffer against poor physical health. However, as perceived stress levels increase, successful use of self-defeating humor for social support reasons may start to fail due to over use of this negative style of humor alienating others, thus leading to an increase in poor health. High levels of self-defeating humor are also associated with high levels of neuroticism and low levels of self-worth. It is possible that an increase in stress brings about an increase in anxiety and neuroticism which exacerbates poor self-worth, all of which are associated with poorer health outcomes (Kuiper & Nicholl, 2004).
Methodological Limitations of the Study
A limitation of this study lies in the sampling method. Employing nonrandom (convenience) sampling techniques is often associated with the resulting sample not being representative of the general population. As the population used in this study consisted of university students, it may not be possible to generalize the results to other populations.
Second, the use of self-report questionnaires does raise questions regarding the reliability and validity of the information gathered. The measurement instruments were selected on the basis that they had been shown to have satisfactory psychometric properties. In the current study, however, all the instruments, except for the HSQ, yielded satisfactory internal consistency. The low Cronbach’s alphas found for the HSQ and its subscales are of concern. Despite evidence of satisfactory internal consistency of the HSQ (e.g., Abel, 2002; Cann & Etzel, 2008; Martin et al., 2003), this study found the Cronbach’s alphas of the HSQ and its subscales to be low. A possible reason for this may be that humor could be culture-based and, as a result, the four humor styles might manifest in different ways between cultural groups (Hiranandani & Yue, 2014). Thus, as this study consisted of four ethnic groups, low internal consistency might have resulted as a consequence of the way these different groups conceptualize humor.
The use of self-report questionnaires can also influence findings because of common-method variance, especially in cross-sectional studies (Lindell & Whitney, 2001). A significant limitation is that cross-sectional designs cannot be used to analyze behavior over time (longitudinally), nor does it allow the researcher to measure the incidence of problems that may develop with time, such as health issues.
Last, although the effect size of the current study indicates some practical value, the small variance explained by the interaction may indicate that other factors could play a more substantial moderating role in the stress–health relationship than humor.
Recommendations for Future Studies
Given the low Cronbach’s alphas regarding the HSQ and its subscales found in this study, it might be beneficial for future studies to focus on the development of a South African–based form of the questionnaire. Future research might benefit from alternative or additional means of assessment, other than the use of self-report measures, such as longitudinal observation or physiological measures. This might enable the researcher to make use of a smaller sample allowing him or her to engage in a more in-depth exploration of variables, focusing on the use of humor in different settings and circumstances. Broadening the scope in future studies to include different variables that have known associations with humor, such as personality traits and different types of stressors, could provide significant contributions to the area of humor research.
Footnotes
Authors’ Note
Both the authors contributed equally.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.