Personality and Individual Differences 48 (2010) 563567

Contents lists available at ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier.com/locate/paid

Individual differences in personal humor styles: Identication of prominent patterns and their associates
Graeme Galloway *
School of Psychological Science, La Trobe University, Bendigo Campus, P.O. Box 199, Bendigo, Victoria 3552, Australia

a r t i c l e

i n f o

a b s t r a c t
Considerable recent research attention is evident in the humor and related literature concerning the topic of humor styles. The major focus to date has been on such styles considered individually rather than the combination of styles characteristic of any given person. Information about the latter topic can arguably advance understanding about humor styles and is examined in the present study. A cluster analysis of participants based on their self-ratings regarding Aggressive, Self-defeating, Afliative, and Self-enhancing humor styles was performed. Four clusters of people were identied consisting of those who score: (1) above average on all of the styles, or (2) below average on all of the styles, or (3) above average on the positive styles (Afliative and Self-enhancing), and below average on the negative styles (Aggressive and Self-defeating), or (4) above average on the negative styles and below average on the positive styles. Big Five personality and Self-esteem correlates of each cluster are reported and the theoretical, practical, and research implications of the ndings are discussed. 2009 Elsevier Ltd. All rights reserved.

Article history: Received 17 December 2008 Received in revised form 6 December 2009 Accepted 10 December 2009 Available online 8 January 2010 Keywords: Humor styles Cluster analysis Psychological health Base arousal Personality

1. Introduction This study concerns the humor use component of sense of humor, the specic aim being to identify and clarify the nature of distinct clusters of people classied by reference to their proles on the Humor Styles Questionnaire (HSQ) (Martin, Puhlik-Doris, Larsen, Gray, & Weir, 2003). The HSQ provides measures of four styles, two characterized by Martin et al. as related positively to psychosocial health and well-being (Afliative and Self-enhancing), and two (Aggressive and Self-defeating) which they describe as negatively related to those variables (p. 72). Two classes of study concerning the HSQ styles are evident in the literature to date examinations of correlations between the styles, and investigations of which other individual difference variables are correlated with each style. Martin et al. (2003) report that the two positive styles are positively correlated, as are the two negative ones. Each of the positive styles are also negatively correlated with depression and anxiety, and positively correlated with Selfesteem, Extraversion, Openness (Martin et al., 2003) and Agreeableness (Saroglou & Scariot, 2002). Both negative styles are negatively correlated with Agreeableness and Conscientiousness, as well as being positively correlated with Neuroticism, hostility, and aggression (Martin et al., 2003). However, simple correlations between the styles cannot provide an accurate indication about the important topic of individual differ* Tel.: +61 3 5444 7511; fax: +61 3 5444 850. E-mail address: g.galloway@latrobe.edu.au 0191-8869/$ - see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.paid.2009.12.007

ences in style proles that is, the styles characteristic of distinct groups of people. Specication of such differences can potentially enhance understanding about the HSQ styles. For example, it is possible that similar scores on the same styles might not be motivated by the same variables for distinct groups of people. As one illustration, people who score above average on the two negative styles might show that pattern because they are lower in psychosocial health and well-being. However, other people who score above average on the two negative styles may do so because they are Extraverts. Klein and Kuiper (2006) indicate that both of the positive and both of the negative styles can assist a person in gaining acceptance in a group which is widely held to be important for Extraverts (see, for example, Donnellan, Oswald, Baird, & Lucas, 2006). The present study aims to add to knowledge about the HSQ styles through examination of the psychological characteristics of the people associated with a distinct style pattern. Now, Martin et al. (2003) hypothesized that the HSQ scales will be related to the Big Five personality variables Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism, and to Self-esteem (pp. 5154). Accordingly, those traits are examined here to assist in interpretation of the style patterns obtained. In terms of these variables, for instance, the abovementioned possibility would be supported by the observation that one group of people who score above average on each HSQ scale score above average on Extraversion, while a different group who score above average on just the negative styles score below average on such things as Self-esteem. Identication of style patterns can be accomplished by cluster analysis which is appropriate to use where the aim is to categorize

564

G. Galloway / Personality and Individual Differences 48 (2010) 563567

people who are similar in some regard (Diekhoff, 1992, pp. 361, 377). The widely used k-means cluster analysis (Quickcluster) is employed to that end in the present study. Application of that technique requires the researcher to specify the number of clusters to be generated and produces discrete groups that are usually easy to interpret. Large numbers of cases can be analyzed using Quickcluster which maximizes variability between clusters, and minimizes variability within clusters, which are requirements of a good cluster solution (Hill & Lewicki, 2007). The technique yields meaningful and useful results in research comparable to that undertaken here (for example, Galloway, 2002). However, the investigator will typically have to estimate how many clusters it is best to extract from the data. A degree of trial and error is therefore usually required in choosing the best cluster solution (see Landau & Everitt, 2004, p. 309), which should be done taking into account the interpretability and parsimony of the solutions considered (Diekhoff, 1992, pp. 368369). There is no one goodness of t test of a given cluster solution (Everitt, Landau, & Leese, 2001). Existing theoretical perspectives on the functions of humor are suggestive about some of the humor style patterns that would make sense if obtained. These include patterns comprising above average or below average scores on each style (see comments on Extraversion above), as well as ones characterized by above average scores on the positive styles and below average scores on the negative ones, and vice versa (see Martin et al., 2003, pp. 5153). Although the Big Five and Self-esteem will be appealed to for interpretation of the patterns of humor styles observed here, it is not suggested that these are the only variables that might motivate the patterns obtained. This issue is examined further in discussion of the present results. 2. Method 2.1. Materials and procedure A survey pack was created consisting of: the International Personality Item Pool (IPIP) which provides measures of each of the Big Five personality constructs (Donnellan et al., 2006); the Rosenberg Self-Esteem Scale (Rosenberg, 1965); and the Humor Styles Questionnaire (Martin et al., 2003). Each of these measures show acceptable levels of reliability. An information statement as well as a demographics data collection sheet were also included in the survey pack. Participants completed and returned the surveys anonymously. 2.2. Participants A convenience sample of 318 Australian participants (mean age = 36.1 years, SD = 14.7, range = 1871) returned usable survey packs, which constitutes a response rate of around 65%. The sample comprises 181 females (mean age = 36.1 years, SD = 14.8, range = 1868), 133 males (mean age = 36.2, SD = 14.7, range = 1871), and four respondents who did not indicate gender. Approximately half of the total sample consists of undergraduate or postgraduate university students. The remainder of the participants were chosen from the general community through networks available to the researcher. 2.3. Data analysis As indicated above, k-means cluster analysis (Quickcluster) was used to categorize respondents on the basis of their scores on each of the HSQ styles. Cronbach alphas and mean inter-item correlations were calculated to check the reliability of the abovementioned scales. Pearson correlations and v2 analysis were used to

examine other relationships of interest. All statistical procedures were performed using SPSS 14.0 for Windows. 3. Results 3.1. Reliability checks It is widely assumed that alpha values greater than .7 are required for acceptable reliability (Nunnally, 1978). The alpha values observed for the Big Five personality variables and for Self-esteem in this study (see Table 2) all exceed that value. Table 1 reports the alpha values for each of the humor style measures. The alphas for the Self-enhancing and the Self-defeating humor styles are very close to .70, but those for Aggressive and for Afliative humor styles are around .60. However, Aron and Aron (2003, p. 607) claim that alphas of .6 or greater are acceptable. In fact, Schmitt (1996) suggests that even a scale with an alpha as low as around .5 may be useful if it has other desirable properties, including meaningful content with respect to the domain covered, and reasonable unidimensionality. In addition, Clark and Watson (1995, p. 316) suggest that the average inter-item correlation is a better measure than coefcient alpha of a scales internal consistency, and recommend values should fall within the range .15.50. Mean inter-item correlations observed in the present study for each humor style scale are: Afliative = .17, Aggressive = .16, Self-defeating = .21, Selfenhancing = .21. The lower alpha values observed for two of the humor scales were therefore not viewed as posing difculties for this research. 3.2. Associates of the humor styles The correlations between the HSQ styles, and between the styles and other individual difference variables, are reported in Tables 1 and 2 respectively. These are provided in order to enable comparison of the ndings observed here with those reported in previous studies. This in turn allows a determination of whether the HSQ provides valid measures for the participants tested in the present study. Several of the associations between the humor styles, and between the styles and other individual difference measures, reect those reported in previous research. Specically, correlations between the humor styles for the total sample (see Table 1) mirror those reported by Martin et al. (2003). Table 2 reports the associations of the four HSQ styles with Self-esteem and the Big Five personality variables. The ndings for the Afliative and Self-defeating styles reect those reported by Martin et al. (2003). For the Aggressive humor style, in addition to the signicant associations observed by Martin et al. (2003), the present results also identied a signicant positive correlation with Extraversion (an association also reported by Saroglou & Scariot, 2002, p. 49), and a signicant negative correlation with Self-esteem. Self-enhancing humor style was observed to be associated with all but one (Openness) of the variables it was found to be associated with by Martin et al. The similarities between the patterns of associations observed here for the humor styles and those

Table 1 Correlations between the humor styles total sample. Humor styles 1. 2. 3. 4. Afliative Aggressive Self-enhancing Self-defeating 1 62 .21* (N = 305) .43* (N = 305) .10(N = 307) 2 .61 .04 (N = 301) .36* (N = 303) 3 4

.69 .06(N = 305)

.68

Cronbach alphas for each scale are included on the diagonal and are all based on N = 318. * The correlation is signicant, p < .05.

G. Galloway / Personality and Individual Differences 48 (2010) 563567 Table 2 Correlations of the humor styles with Self-esteem and Big Five personality variables. Humor styles Alphas = Afliative Ns Aggressive Ns Self-enhancing Ns Self-defeating Ns
**

565

Table 4 Mean z scores by cluster for Self-esteem and Big Five personality variables. Clusters* 1 Self-esteem Openness Conscientiousness Extraversion Agreeableness Neuroticism Ns for means
*

SE .84 .21* 305 .17* 301 36* 303 .38* 306

O .70 .27* 306 .08 303 .07 303 .03 306

C .78 .12 310 .35* 308 .02 307 .28* 309

E .82 .50* 309 .16* 305 .29* 306 .04 309

A .71 .06 310 .43* 307 .16* 307 .16* 310

N .78 .08 310 .14* 307 .31* 307 .36* 310

2 .12 .09 .39 .35 .05 .02 61

3 .50 .01 .27 .23 .55 .45 73

4 .30 .27 .11 .45 .35 .12 56

.01 .23 .33 .43 .32 .10 93

SE = Self-esteem; O = Openness; C=Conscientiousness; E = Extraversion; A = Agreeableness; N = Neuroticism. * The correlation is signicant, p < .05. ** Cronbach alphas are all based on N = 318.

See Table 3 for description of clusters.

Table 5 Observed and expected percentages of males and females in each cluster. Clusters 1 Females (N = 168) Number in each cluster 47 Expected% of females in each cluster = 57% Observed % 49% Males (N = 127) Number in each cluster 49 Expected% of males in each cluster = 43% Observed% 51% 2 44 67% 22 33% 3 48 65% 26 35% 4 29 49% 30 51%

obtained in other studies are consistent with the suggestion that the HSQ does provide a valid measure of humor style for the participants in this study. 3.3. The cluster analysis As indicated above, k-means cluster analysis was used to identify categories of people characterized by distinct patterns of scores on the HSQ scales. Participants mean raw scores on each of the four humor styles and on each of the Big Five personality variables as well as Self-esteem were transformed into z scores to facilitate interpretation of the results. Several k-means cluster analyses were conducted, with a four-cluster solution judged to be the most appropriate. Specically, as indicated below, each of the four clusters are easily interpretable within existing theoretical frameworks. The four-cluster solution is also more parsimonious, given that ve and six cluster solutions which were also examined resulted in a replication of, respectively, one or two of the clusters in the four-cluster solution. As indicated above, interpretability and parsimony are criteria for deciding on the best cluster solution. The mean z scores on each humor style by cluster and the percentage of participants in each cluster are reported in Table 3. Cluster 1 comprises people who scored above average on each of the humor styles, those in Cluster 2 are below average on each style, Cluster 3 are above average on the positive styles, and below average on the negative styles, and Cluster 4 are above average on the negative styles and below average on the positive ones. Table 4 reports mean z scores on Self-esteem and the personality traits for the people in each cluster. Members of Cluster 1 are above average on Openness and Extraversion, and below average on Conscientiousness and Agreeableness, but are close to the mean on Self-esteem and Neuroticism. Members of Cluster 2 are above

average on Conscientiousness, below average on Extraversion, and close to the mean on the other traits. Cluster 3 members are above average on Self-esteem, Conscientiousness, Extraversion, and Agreeableness. They are also below average on Neuroticism, and close to the mean on Openness. Members of Cluster 4 are below average on Self-esteem, Openness, Extraversion, and Agreeableness, and close to the mean on Conscientiousness and Neuroticism. The number of males and females in each cluster was examined in order to assist interpretation of the patterns of styles identied (see Table 5). A signicant chi square value (v2 = 8.4, df = 3, p < .05) is associated with these numbers indicating differences between clusters regarding the expected and observed proportions of males and females in each. Specically, Clusters 1 and 4 comprise a larger than expected proportion of males, and Clusters 2 and 3 contain a greater proportion of females than expected. These results are examined in more detail below. 4. Discussion 4.1. The nature of the clusters The personality and gender-based proles of each cluster of people identied here are suggestive about possible determinants of the style patterns observed. Specically, the members of Cluster 1 (above average on each style) are outgoing and open to new experiences. They are also impulsive given that they scored below average on Agreeableness and Conscientiousness which are negatively related to Impulsivity (Acton, 2003). Members of Cluster 2 (below average on each style) are restrained, not outgoing, as well as being focused and organized. A variable that might underlie the patterns of styles observed for Clusters 1 and 2 is base arousal. Galloway and Chirico (2008) indicate that people who are higher in base arousal, for instance, higher Neuroticism scorers, do not enjoy stimulation as much as do people who are lower in base arousal as, for instance, higher sensation seekers are assumed to be. Several indicators of low base arousal characterize the members of Cluster 1. Specically, they

Table 3 Mean z scores on each humor style for each cluster and percentage of participants in each cluster. Clusters* Ns Total N = 296 % of total N Humor styles Self-enhancing Self-defeating Aggressive Afliative
*

1 97 33%

2 66 22%

3 74 25%

4 59 20%

.40 .66 .82 .75

1.13 .60 .58 .69

.68 .74 .82 .52

.27 .47 .32 .94

Cluster 1 = above average on each style; Cluster 2 = below average on each style; Cluster 3 = above average on the positive styles, below average on the negative styles; Cluster 4 = above average on the negative styles, below average on the positive styles.

566

G. Galloway / Personality and Individual Differences 48 (2010) 563567

scored above average on Extraversion which is positively associated with sensation seeking (Eysenck & Zuckerman, 1978). Higher Impulsivity is also associated with liking of arousal (Acton, 2003), as is higher Openness (Arnett, 1994). Furthermore, as reported in Table 5, Cluster 1 contains a greater than expected proportion of males, males tending to like arousal more than do females (Galloway, Mitchell, Getz, Crouch, & Ong, 2008). Now, as indicated earlier in this paper, each humor style can facilitate social interaction which in turn is a source of arousal. Furthermore, arousal theories of humor propose that it involves an increase in, and then reduction of, arousal/tension (see Martin, 2007). Given this, the above average scores on each humor scale characteristic of Cluster 1 can be viewed as consistent with a desire of under-aroused individuals to increase their arousal levels. On the other hand, the above average scores on Conscientiousness and below average scores on Extraversion associated with members of Cluster 2 are consistent with them being higher in base arousal (Revelle & Loftus, 1990). Cluster 2 also contains a greater than expected percentage of females, and females tend to like arousal less than do males (Galloway et al., 2008). The below average scores for members of Cluster 2 on each of the HSQ scales are therefore consistent with a desire to avoid increasing an already high base arousal. The members of Cluster 3 (above average on the positive styles, below average on the negatives) can be portrayed as well-balanced, low in anxiety, and positive towards themselves and others. Such people enjoy higher status in groups and other interpersonal contexts (see Robins, Hendin, & Trzesniewski, 2001, p. 160), and are motivated to maintain or enhance positive feelings about themselves (Burton, Westen, & Kowalski, 2009). Given this, their above average scores on the positive humor scales are not surprising. The positive styles can reinforce their already positive relationships with those who afford them higher social status. This could involve use of more lighthearted humor content, such as satire, irony, and philosophical humor. The people in Cluster 4 (above average on the negative styles, below average on the positives) can be described as not being open to new experiences as well as being negative towards themselves and others. Such people might use humor to defend against perceived threats to their self-image. Specically, their use of aggressive humor can devalue those who they perceive as not valuing them. This could result in humor content which is sexist, racist, or sick (see Long & Graesser, 1988, for taxonomies of jokes and wit). Also, as noted by Martin et al. (2003), self-defeating humor might help such individuals deny to themselves and others that they harbor feelings of low self-worth. The analysis provided above extends knowledge about the HSQ scales beyond what can be known about them through examination of correlates of the individual scales. Specically, simple correlations cannot indicate the number of distinct style patterns that characterize a given cohort of people, nor the combination of individual difference traits which motivate any such pattern. In fact, simple correlations can be misleading about the associates of humor styles for different people. Consider the following examples based on simple correlations reported in Table 2. The two positive styles as well as the Aggressive style are positively correlated individually with Extraversion. The people in Cluster 1, and those in Cluster 3, are above average on Extraversion. However, members of Cluster 1 are above average on all of the styles, while those in Cluster 3 are above average on the positives and below average on the negative styles. Self-esteem is positively correlated with the positive styles, and negatively correlated with the negative ones. Although such relationships are mirrored in Clusters 3 and 4, Self-esteem is not related to any of the styles for people in Clusters 1 and 2. Agreeableness is signicantly positively correlated individually with Self-enhancing style. However, for people in

Cluster 1, below average scores on Agreeableness are associated with above average scores on that scale, whereas for Cluster 3 above average scores on Agreeableness go together with above average scores on the Self-enhancing scale. Openness is not correlated with three of the four styles considered individually, but for Clusters 1 and 4 Openness is related to all of the styles. Several issues requiring further research regarding this topic will now be considered. 5. Suggestions for further research 5.1. Replication of the patterns of humor styles and their associates An obvious issue for further research concerns whether the patterns of humor styles and their individual difference correlates observed here replicate for samples similar to the present one. Replication of the results would serve to increase condence in the reliability of the patterns observed. Another important issue concerns whether the present patterns replicate in cultures different from the Australian one and, if not, which aspects of a given culture inuence the style proles of its members. 5.2. Effects and associates of patterns of humor styles Another issue for future research concerns the effects particular patterns of humor styles bring about. For instance, the aggressive style characteristic of people in Cluster 4 (who are not open to new experiences as well as being negative towards themselves and others) might reduce perceived threats to their self-image. However, such humor might also occasion a negative reaction towards the user from those it is directed at which might reinforce the users lower self-image. The self-defeating style associated with Cluster 4 might also negatively inuence self-image because it involves self-criticism and emotional repression (Martin et al., 2003). The below average scores on each of the HSQ scales associated with individuals who are restrained, highly focused and organized (Cluster 2) could in some contexts have the desired effect of lessening interactions with others. However, this could also give rise to anxiety-arousing negative reactions from other people. Examination of such issues would best be carried out using a longitudinal research design (see Galloway & Cropley, 1999). Furthermore, the personality links identied here for the people associated with a particular humor style pattern also provide a basis for suggestions about some of the other variables it might be worth investigating in order to increase knowledge about the motivators of such patterns. For instance, each of the personality variables members of Clusters 1, 2, 3, and 4 score above or below the mean on are related to one or more of the variables locus of control, generalized self-efcacy, self-monitoring (Barrick, Parks, & Mount, 2005; Graziano & Ward, 1992; Judge, Erez, Bono, & Thoresen, 2002), and sensation seeking (Eysenck & Zuckerman, 1978). Of course, it is an empirical question for further research whether above or below average scores on the personality traits examined in this study are associated with higher or lower mean scores on other variables the former are correlated with individually. The observation that such is the case for some or all of the abovementioned variables will result in greater understanding about the motivators of style patterns and could be useful regarding such things as how to address problems associated with a particular style prole. For instance, as indicated above, the style pattern characteristic of Cluster 4 could occasion detrimental effects for members of that cluster. Now, relationships reported in the abovementioned studies raise the possibility that members of Cluster 4 might be below average on self-monitoring and generalized self-efcacy, and more

G. Galloway / Personality and Individual Differences 48 (2010) 563567

567

external locus of control. Lower self-monitors appear to be less able than are higher self-monitors to use feedback from others as a basis to manage their own behaviour (Lammers, 1991). People who are lower in generalized self-efcacy tend to feel that they cannot perform well in various situations, and those characterized as more external locus of control are more likely to believe that other people and events are primarily responsible for the positive and negative outcomes of their behaviour (Judge et al., 2002). If those features do indeed characterize people who score above average on the negative styles and below average on the positives then assisting them to adopt more adaptive attitudes and behaviours might reduce their propensity to use negative humor styles when interacting with others. Relevant to that aim is the observation that people who are higher in external locus of control are more likely than are internals to change their attitudes when presented with a compelling message (Judge et al., 2002). As also noted above, people who score below average on each HSQ scale (Cluster 2 in this study) could occasion a negative reaction from other people in some situations. The abovementioned research raises the possibility that people with the personality prole identied here for the members of Cluster 2 might be low self-monitors. If that turns out to be the case, then assisting them to take into account the feedback from others about their behaviour might help them to be more aware of and avoid the negative effects of their style prole. 6. Conclusions The present study has increased knowledge about the nature of the HSQ humor styles beyond what can be known about them through examination of simple correlations between individual styles and other variables by identifying the distinct style patterns that characterize different people and some personality traits which appear to motivate such patterns. A key nding of this research is that simple correlations are in many cases uninformative about how the humor styles and personality will be related for different people. Further productive research on this topic should examine the replicability of the style proles identied here in different cultures as well as which other personality and individual difference variables motivate distinct humor style proles. Such information could in turn have practical applications regarding such things as how best to address difculties associated with particular style patterns. References
Acton, G. (2003). Measurement of impulsivity in a hierarchical model of personality traits: Implications for substance use. Substance Use & Misuse, 38(1), 6783. Arnett, J. (1994). Sensation seeking: A new conceptualization and a new scale. Personality and Individual Differences, 16(2), 289296.

Aron, A., & Aron, E. (2003). Statistics for psychology (3rd ed.). Upper Saddle River, NJ: Prentice-Hall. Barrick, M., Parks, L., & Mount, M. (2005). Self-monitoring as a moderator of the relationships between personality traits and performance. Personnel Psychology, 58, 745767. Burton, L., Westen, D., & Kowalski, R. (2009). Psychology (2nd ed.). Milton, Qld: Wiley. Clark, L., & Watson, D. (1995). Constructing validity: Basic issues in objective scale development. Psychological Assessment, 7(3), 309319. Diekhoff, G. (1992). Statistics for the social and behavioral sciences. Dubuque, IA: Brown. Donnellan, M. B., Oswald, F., Baird, B., & Lucas, R. (2006). The Mini-IPIP: Tiny yet effective measures of the Big Five factors of personality. Psychological Assessment, 18(2), 192203. Everitt, B., Landau, S., & Leese, M. (2001). Cluster analysis (4th ed.). London: Arnold. Eysenck, S., & Zuckerman, M. (1978). The relationship between sensation seeking and Eysencks dimensions of personality. British Journal of Psychology, 69(4), 483487. Galloway, G. (2002). Psychographic segmentation of park visitor markets: Evidence for the utility of sensation seeking. Tourism Management, 23(6), 581596. Galloway, G., & Chirico, D. (2008). Personality and humor appreciation: Evidence of an association between trait neuroticism and preferences for structural features of humor. Humor, 21(2), 129142. Galloway, G., & Cropley, A. (1999). Benets of humor for mental health: Empirical ndings and directions for further research. Humor, 12(3), 301314. Galloway, G., Mitchell, R., Getz, D., Crouch, G., & Ong, B. (2008). Sensation seeking and the prediction of attitudes and behaviours of wine tourists. Tourism Management, 29(5), 950966. Graziano, W., & Ward, D. (1992). Probing the Big Five in adolescence: Personality and adjustment during a developmental transition. Journal of Personality, 60(2), 425439. Hill, T., & Lewicki, P. (2007). Statistics: Methods and applications. Tulsa, OK: Statsoft. Judge, T., Erez, A., Bono, J., & Thoresen, C. (2002). Are measures of self-esteem, neuroticism, locus of control, and generalized self-efcacy indicators of a common core construct? Journal of Personality and Social Psychology, 83(3), 693710. Klein, D., & Kuiper, N. (2006). Humor styles, peer relationships, and bullying in middle childhood. Humor, 19(4), 383404. Lammers, H. B. (1991). Moderating inuence of self-monitoring and gender on responses to humorous advertising. The Journal of Social Psychology, 131(1), 5769. Landau, S., & Everitt, B. (2004). A handbook of statistical analysis using SPSS. Boca Raton, Fl: CRC Press. Long, D., & Graesser, A. (1988). Wit and humor in discourse processing. Discourse Processes, 11, 3560. Martin, R. (2007). The psychology of humor: An integrative approach. Burlington, MA: Elsevier Academic Press. Martin, R., Puhlik-Doris, P., Larsen, G., Gray, J., & Weir, K. (2003). Individual differences in uses of humor and their relation to psychological well-being: Development of the Humor Styles Questionnaire. Journal of Research in Personality, 37, 4875. Nunnally, J. C. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill. Revelle, W., & Loftus, D. (1990). Individual differences and arousal: Implications for the study of mood and memory. Cognition and Emotion, 4(3), 209237. Robins, R., Hendin, H., & Trzesniewski, K. (2001). Measuring global self-esteem: Construct validation of a single-item measure and the Rosenberg self-esteem scale. Personality and Social Psychology Bulletin, 27(2), 151161. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Saroglou, V., & Scariot, C. (2002). Humor Styles Questionnaire: Personality and educational correlates in Belgian high school and college students. European Journal of Personality, 16, 4354. Schmitt, N. (1996). Uses and abuses of coefcient alpha. Psychological Assessment, 8(4), 350353.