Multidimensional Sense of Humor Scale

Zurich Open Repository and
Archive
University of Zurich
University Library
Strickhofstrasse 39
CH-8057 Zurich
www.zora.uzh.ch
Year: 2018
Psychometric evaluation of the revised Sense of Humor Scale and the construction
of a parallel form
Ruch, Willibald ; Heintz, Sonja
DOI: https://doi.org/10.1515/humor-2016-0085
Posted at the Zurich Open Repository and Archive, University of Zurich

ZORA URL: https://doi.org/10.5167/uzh-150295
Journal Article
Published Version
Originally published at:

Ruch, Willibald; Heintz, Sonja (2018). Psychometric evaluation of the revised Sense of Humor Scale and the
construction of a parallel form. HUMOR: International Journal of Humor Research, 31(2):235-257.
DOI: https://doi.org/10.1515/humor-2016-0085
Humor 2018; 31(2): 235–257
Willibald Ruch* and Sonja Heintz

Psychometric evaluation of the revised
Sense of Humor Scale and the construction
of a parallel form
https://doi.org/10.1515/humor-2016-0085
Abstract: McGhee (1996, Health, healing and the amuse system: Humor as survi-
val training. Dubuque: Kendall/Hunt; 1999, Health, healing and the amuse sys-
tem: Humor as survival training (3rd edition). Dubuque: Kendall/Hunt) proposed
a model of the sense of humor including the six “humor skills” of enjoyment of
humor, laughter, verbal humor, finding humor in everyday life, laughing at
yourself, and humor under stress, measured with the Sense of Humor Scale
(SHS). The purpose of the present study is to evaluate the psychometric proper-
ties of the SHS (revised version from 1999) and to develop a parallel form of the
SHS to double the amount of items for each humor skill. Combing these two
forms should yield reliable and factorially valid scales of the six humor skills.
Participants in two online studies (n = 315 and 542) completed the SHS and its
parallel form, along with measures of various outcomes. The psychometric
properties of the SHS were of mixed quality, and those of the parallel form
were uniformly good. The parallel-test reliability was sufficiently high to regard
the two scales as parallel versions. Combining the two measures resulted in
reliable and distinguishable scales of the six humor skills. All humor skills
correlated positively with humor-related attitude and mood, cheerfulness, and
life satisfaction. Importantly, they spanned different dimensions of the sense of
humor, underscoring the usefulness of each humor skill.
Keywords: Sense of Humor Scale, psychometrics, sense of humor model, humor

training, humor skills
1 Introduction
In 1996, McGhee proposed a multi-faceted concept of the sense of humor, an
assessment tool for its measurement, and a training program for enhancing the
*Corresponding author: Willibald Ruch, Department of Psychology, University of Zurich, Zurich,

Switzerland, E-mail: [email protected]
Sonja Heintz, Department of Psychology, University of Zurich, Zurich, Switzerland,
E-mail: [email protected]
Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

Authenticated | [email protected] author's copy
Download Date | 5/2/18 5:07 PM
236 Willibald Ruch and Sonja Heintz
sense of humor. McGhee (1979) stated that humor is a form of play – the play
with ideas. Without a playful frame of mind, the same event may be seen as
interesting, puzzling, annoying, frightening, etc., but not as funny. Playfulness
and seriousness were assigned core roles in the sense of humor model. While
people in general might be capable of spotting incongruities, absurdities, and
ironies of life, only the mentally playful will find humor in them. A habitually
serious attitude or frame of mind will hinder treating incongruities humorously.
McGhee (1996) postulated that playfulness is inherited and that all children
play, but that socialization (entering school, the workplace) may counteract
this tendency to play and may cause individuals to lose their ability to be
playful. The humor-training program for the “terminally serious” aims at redis-
covering a playful attitude or outlook on life with the claim that rediscovery is a
key element for positive changes (McGhee 1996; Ruch and McGhee 2014).
While playfulness is the foundation of the sense of humor, it is not a
quality specific to humor. McGhee (1996) proposed six genuine humor skills,
namely: enjoyment of humor, laughter, verbal humor, finding humor in every-
day life, laughing at yourself, and humor under stress. In the first version of
the model, these six elements were hierarchically organized from enjoyment of
humor to finding humor under stress, inasmuch as the ones higher in the
hierarchy were assumed to be more difficult to develop. To measure these
skills, McGhee (1996) introduced the sense of humor scale (SHS), which later
(McGhee 1999) underwent a revision. [This revised version is also printed in
McGhee (2010).] Both versions of the SHS were deductively developed. They
differ with respect to the number of items (five vs. four per humor skill) and the
answer format of the response scale (four-point vs. seven-point, 1 = strongly
disagree, 4/7 = strongly agree). The six humor skills are the same in both
versions and may be added together to form a “total sense of humor” (SoH)
score or a “humor quotient”. In the first version of the SHS, this score was also
weighted (i. e., laughing at yourself weighted by a factor of 1.5 and humor
under stress weighted by 2).
A first psychometric evaluation (Ruch and Carrell 1998), which investigated
the SHS along with McGhee’s scales of the foundation of humor (i. e., mental
frame of mind and mood quality) yielded high internal consistencies for an
overall score (SHS and two scales of “seriousness and negative mood” and
“playfulness and positive mood”) in both an American and a German sample
(0.90 to 0.92). However, the five-item skills laughter and enjoyment of humor
yielded Cronbach’s alphas lower than 0.70 (see also Wrench and McCroskey
2001). Furthermore, it was more fruitful to build bipolar scales of positive vs.
negative mood and playful vs. serious attitude. The revision of the scale was
partly successful, as investigated mostly with the German version of the SHS.

Sense of Humor Scale 237
The number of items for the SHS humor skills was dropped from five to four
without a substantial loss of reliability for four of the scales; however, enjoy-
ment of humor and laughter again yielded coefficients below 0.70 (Proyer et al.
2010; only for enjoyment of humor in Proyer 2014). Positive vs. negative mood
and playfulness vs. seriousness yielded satisfactory internal consistencies
(Müller and Ruch 2011; Proyer et al. 2010, 2012; Proyer 2014). Also the SoH
score and laughing at yourself proved to be internally consistent in several
studies (Beermann and Ruch 2011; Proyer et al. 2012; Proyer 2014; Ruch and
Heintz 2016).
Several studies investigated the validity of the SHS, but some essential
information is still missing. A principal component analysis of the eight scales
(positive and negative mood/attitude and the SHS) in German and English
samples yielded separate but highly correlated factors for the six humor skills
and the playful vs. serious dimensions (Ruch and Carrell 1998). However, no
factor analysis at the item level has been undertaken yet to see whether the six
concepts can be distinguished. Wrench and McCroskey (2001) conducted a factor
analysis of 48 items (24 SHS items and 24 inverted items) and it “[…] revealed a
strong primary factor” (p. 148), suggesting the use of the total SoH score only.
Thus, it yet remains to be demonstrated that the six humor skills can be
recovered as separate factors.
There is ample evidence that the SHS correlated very highly (0.60– 0.74)
with other humor scales. For example, the SoH score correlated with the
VIA-IS humor scale, the self- and peer rated socially warm (vs. socially cold)
style of humorous conduct, and trait cheerfulness (Müller and Ruch 2011;
Ruch and Carrell 1998; Ruch et al. 2011; Wrench and McCroskey 2001). In
fact, all facets of trait cheerfulness and the six humor skills belonged to the
same factor, suggesting these might be almost interchangeable concepts
(Ruch and Carrell 1998). This was also found for the revised SHS, in which
all scales and the SoH score loaded strongly on a cheerful-engaged playful-
ness factor (Proyer 2014).
Soury and Devillers (2014) found that the SoH score correlated positively
with the frequency and length of recorded laughter in response to four passive
and active laughter-inducing tasks (such as watching funny videos or a tongue-
twister game). Beermann and Ruch (2011) supported the convergent validity of
laughing at yourself with a moderate convergence between self-reports, peer-
reports, and the frequency of observed smiling and laughter in response to
distorted portraits of oneself (see Hofmann, this issue).
Only indirect evidence exists for the idea that there is a hierarchy in the
humor skills. Humor under stress had the lowest mean and enjoyment of humor
had the highest mean, but the proposed rank order was not found for all scales

(Ruch and Carrell 1998). However, Müller and Ruch (2011) found that both
competent (vs. inept) and reflective (vs. boorish) humor styles tended to corre-
late most highly with the components of the sense of humor that are most
difficult to acquire and low to negative with the ones that are easiest to develop.
The ultimate test would be the latency needed to develop these humor skills in a
training setting; this would clarify the best order among the concepts. Similarly,
it is not known whether playfulness (and low seriousness) are “motors” for the
humor skills.
The combination of the six humor skills into the SoH score in both theory
and empirical research is reflective of the early notion that sense of humor is
unidimensional. The SHS was developed when multi-dimensional measure-
ments and models were as yet unknown, as, for instance, the State-Trait-
Cheerfulness-Inventory (STCI; Ruch et al. 1996) and the Humorous Behavior Q-
Sort Deck (Craik et al. 1996). Thus, locating the individual humor skills in a
multidimensional humor space is of interest. A recent multidimensional model
of the sense of humor (Ruch 2012a) was derived from factor analyses of self- and
peer-reports of the Humorous Behavior Q-Sort Deck (Craik et al. 1996). This
model of four sense of humor dimensions entails social fun (being a funny
entertainer and showing prosocial humor in groups), mockery (entailing sar-
casm, cynicism, laughing at others, making fun of others, and criticizing others
humorously), humor ineptness (not being able to tell jokes or to laugh at
oneself, and reacting inappropriately to humor and jokes), and cognitive/reflec-
tive humor (making witty comments and word plays, liking intellectually and
sophisticated humor).
1.1 Aims of the present study
The present study has five aims. First, we test the psychometric properties of
the revised SHS (McGhee 1999): specifically, the reliability (internal consis-
tency and unidimensionality) of the six scales and the SoH score and the
corrected-item total correlations (CITCs) of the 24 items. Second, we develop
and psychometrically test a parallel form of the SHS (SHS-P), a form that also
entails four items per humor skill. We expect the SHS-P to be at least as
psychometrically sound as the SHS. Additionally, the parallel-test reliability
of the humor skills of the SHS and the SHS-P is expected to be sufficiently
high. Although the SHS-P could potentially be used as a stand-alone instru-
ment to measure the six humor skills, we did not intend this usage in the
present study. Instead, our third aim is to combine the items of the SHS and
the SHS-P (resulting in the combined SHS) to allow for a longer and

consequently more reliable and factorially valid measurement of the six humor
skills. Fourth, the overlap of the six humor skills with humor-related measures
(playful vs. serious attitude, positive vs. negative mood, the temperamental
basis of the sense of humor, and four dimensions of the sense of humor) and
life satisfaction is investigated. As attitude and mood were considered the
“motor” of the sense of humor (McGhee 1996, 1999), each humor skill should
positively correlate with playful vs. serious attitude and positive vs. negative
mood. Next, it is tested whether the combined SHS and cheerfulness could be
regarded as more or less interchangeable constructs (as was found by Ruch
and Carrell 1998). Also, we examine whether the humor skills differentially
relate to the other humor-related concepts. Finally, each humor skill and the
SoH score are expected to correlate positively with life satisfaction, in accor-
dance with McGhee’s claims that they would foster positive emotions and
resilience, which should eventually result in enhanced evaluations of one’s
life. Life satisfaction was also deemed relevant because it complements posi-
tive and negative mood (as affective components of subjective well-being) as
the cognitive component of subjective well-being.
2 Methods
2.1 Participants
2.1.1 Sample 1
Overall, 324 participants completed all items of the SHS and the SHS-P. A total of
315 participants (42.9 % men) with a mean age of 36.67 (SD = 14.66) ranging from
18 to 85 years provided valid responses in this study. Participants were primarily
Swiss (67.6 %) or German (18.4 %), but several other nationalities were also
represented (13.9 %). Most participants were well educated, with 53.3 % having
passed tertiary education, 19.0 % having A-levels, 23.4 % having an apprentice-
ship, and 4.4 % having up to nine years of education. Two-thirds (65.7 %) of the
participants were employed.
2.1.2 Sample 2
Overall, 544 participants completed all items of the SHS and the SHS-P. A total
of 542 participants (28.8 % men) with a mean age of 40.45 (SD = 12.86) ranging

from 18 to 76 years provided valid responses. Subsets of this sample also

provided data on the temperamental basis of the sense of humor (n = 177,
30.5 % men, Mage = 42.54), the four sense of humor dimensions (n = 140, 28.6 %
men, Mage = 39.94), and life satisfaction (n = 513, 28.5 % men, Mage = 40.68).
Participants were primarily German (66.2 %), Swiss (20.8 %), and Austrian
(8.5 %), but several other nationalities were also represented (4.5 %).
Participants were rather well educated, with 59.8 % being university students
or having passed tertiary education, 23.3 % having up to nine years of education,
11.3 % having A-levels, and 5.6 % having an apprenticeship. Most participants
(76.0 %) were employed.
To test the psychometric properties of the SHS, the SHS-P, and the combined
SHS, the two samples were analyzed together, resulting in 857 participants
(34.0 % males) with a mean age of 39.06 (SD = 13.66).1
2.2 Instruments
2.2.1 Sense of Humor Scale (SHS; as revised by McGhee 1999)
The SHS comprises 24 items, four each for the six humor skills (enjoyment of
humor, laughter, verbal humor, finding humor in everyday life, laughing at
oneself, and humor under stress). The scale comprises a seven-point Likert-
like answer format from strongly disagree (1) to strongly agree (7). The scale
midpoint (4) is labeled “neutral”.
2.2.2 Parallel form of the Sense of Humor Scale (SHS-P)
The SHS-P was developed by having graduate psychology students familiar with
test construction write 52 items for the six skills based on McGhee’s descriptions,
taking into account the psychometric properties (i. e., CITCs and item intercor-
relations) and wording of the SHS items. Experts on test construction and humor
supervised the students during this process. The items were adapted and fina-
lized in discussions between the experts and the first author, resulting in 24 new
items (four per humor skill). While the item contents should differ from the SHS
items, each humor skill should be parallel to the SHS (i. e., the SHS-P should
measure the same humor skills as the SHS). The scale comprises the same
1 Results were similar when the two samples were analyzed separately.

answer format as the SHS. The SHS-P items can be obtained by contacting the
authors.
2.2.3 Humor-related attitude and mood (McGhee 1999)
In addition to the SHS, McGhee (1999) also proposed two scales that measure
playful vs. serious attitude and positive vs. negative mood with eight items each.
The scales employ a seven-point Likert-like answer format, with varying answer
options (e. g., from very sad [1] to very happy [7]). The scale midpoint (4) is
always labeled “neutral”. Internal consistencies in the present study (N = 857)
were 0.73 for attitude and 0.86 for mood.
2.2.4 State-Trait-Cheerfulness-Inventory, Trait Version<60> (STCI-T<60>; Ruch

et al. 1996)
The STCI-T < 60 > is a 60-item questionnaire measuring trait cheerfulness, ser-
iousness, and bad mood using a four-point Likert scale from strongly disagree (1)
to strongly agree (4). Internal consistencies in the present study (n = 177) were
0.94 (cheerfulness), 0.86 (seriousness), and 0.96 (bad mood).
2.2.5 Four Dimensions of Humor Scale (4DHS; Ruch 2012b)
The 4DHS is a short scale containing markers of the four sense of humor
dimensions (Ruch 2012a) with six items each. Sample items are “I have a
reputation of being a funny entertainer” (social fun), “I have an earthy, salty
humor” (mockery), “I cannot laugh about my own weaknesses and failures”
(humor ineptness), and “I like hearing and creating witty wordplays” (cognitive/
reflective humor). The instrument employs a seven-point Likert-scale from
strongly disagree (1) to strongly agree (7). Internal consistencies in the present
study (n = 140) were 0.79 (social fun), 0.82 (mockery), 0.54 (humor ineptness),
and 0.73 (cognitive/reflective humor).
2.2.6 Satisfaction With Life Scale (SWLS; Diener et al. 1985)
The SWLS assesses general satisfaction with life, employing a seven-point

Likert-scale from strongly disagree (1) to strongly agree (7). Cronbach’s alpha
was 0.86 in the present study (n = 513).

2.3 Procedure
Sample 1 was recruited by several graduate students via personal invitations

(e-mails and online social networks). Participants received no compensation.
Other variables were collected that are not relevant for the present study.
Sample 2 was collected via a website for research purposes (www.char
akterstaerken.org). This website hosts research instruments related to posi-
tive psychology, personality, and humor. It has been promoted by different
means, such as press coverage, publishing the link on different websites, by
sending regular newsletters, and also by contacting particular groups to
obtain heterogeneous samples. Selection criteria for participants are an age
of 18 years and above and a reasonable command of German. Participants
who completed the SHS and the SHS-P, humor-related attitude and mood,
the STCI-T < 60 > , the 4DHS, or the SWLS in 2014 were included in the
present study. Participants received an automated and personalized feedback
after completing each questionnaire. All instruments were administered in
German in both samples.
2.4 Data analysis
Cronbach’s alpha and the mean inter-item correlations served as indicators of

the internal consistencies of the scales. The quality of the items was assessed by
comparing the CITCs with the median correlations with the remaining humor
skills of the SHS and the SHS-P. Unidimensionality of the SHS and SHS-P scales,
parallel-test reliability, and the factor structure of the six combined SHS humor
skills were tested in confirmatory factor analyses (CFA) using the lavaan pack-
age (Rosseel 2012) in R (R Core Team 2015). The Satorra-Bentler’s maximum
likelihood mean–adjusted (MLM) estimator was used in the CFA models. Both fit
indices with (χ2, χ2/df, comparative fit index [CFI], root mean square error of
approximation [RMSEA], and standardized root mean square residual [SRMR])
and without recommended cut-offs (Akaike information criterion [AIC] and
sample-size adjusted Bayesian [BIC], with smaller values indicating a better fit)
were employed to compare different structural models of the combined SHS
scales. The interpretation of the fit indices was based on Schermelleh-Engel et al.
(2003), who recommend for an acceptable fit: χ2 at p > 0.01, χ2/df ≤ 3, CFI ≥
0.95, RMSEA ≤ 0.08, and SRMR ≤ 0.10. Standard multiple regression analyses
were conducted to test the overlap between the combined SHS scales and the
scales of the STCI-T < 60 > and the 4DHS.

3 Results
3.1 Psychometric properties of the SHS and SHS-P scales

and items
3.1.1 Scale-level analyses
Table 1 shows the descriptive statistics of the SHS and SHS-P scales to evaluate
their psychometric properties.
As can been seen in Table 1, Cronbach’s alphas of the SHS scales ranged
from 0.56 (enjoyment of humor) to 0.84 (humor under stress) with a median
Table 1: Descriptive statistics and internal consistencies of the scales of the Sense of Humor
Scale (SHS) and the Parallel Form of the SHS (SHS-P).
Scales Ma SD Sk K Alpha MIIC rgender rage
Enjoyment of humor
SHS . . . −. . . . −.***
SHS-P . . −. . . . −. −.***
Laughter
SHS . . −. . . . . −.**
SHS-P . . −. . . . .*** .
Verbal humor
SHS . . −. −. . . −.*** −.*
SHS-P . . −. . . . −. −.***
Everyday life
SHS . . −. . . . .* −.*
SHS-P . . −. . . . . −.
Laughing at yourself
SHS . . −. . . . . −.
SHS-P . . −. . . . . .
Humor under stress
SHS . . −. −. . . . −.
SHS-P . . −. . . . −. .
Total sense of humor
SHS . . −. . . . . −.**
SHS-P . . −. . . . . −.
Note: N = 857 (n = 856 for age). Everyday life = finding humor in everyday life, Sk = skewness,
K = excess kurtosis, Alpha = Cronbach’s alpha, MIIC = mean inter-item correlation,
rgender = correlation with gender (male = 1, female = 2), rage = correlation with age.*p < 0.05.
**p < 0.01. ***p < 0.001.atheoretical minimum = 1 and maximum = 7.

of 0.80. The scales of the SHS-P showed higher internal consistencies,

ranging from 0.71 (enjoyment of humor) to 0.88 (humor under stress) with
a median of 0.84. The SoH scores had good internal consistencies. Mean
inter-item correlations were numerically larger for the SHS-P (median = 0.54)
than for the SHS (median = 0.42). Gender and age differences were generally
small.
A CFA model with one latent factor for each scale of the SHS and the SHS-P
and the SoH scores tested their unidimensionality (homogeneity). Table 2 shows
the fit indices of the CFA models.
As shown in Table 2, most of the fit indices indicated an acceptable fit for all
scales except for the verbal humor and laughing at yourself SHS scales and both
SoH scales. Thus, for 10 of the 14 scales unidimensionality can be supported.
Importantly, the SoH score was not homogenous, indicating that separating the
humor skills could provide a better fit to the data.
Table 2: Fit indices of the unidimensionality models of the Sense of Humor Scale (SHS) and the
Parallel Form of the SHS (SHS-P) and range of the standardized loadings.
Scales χ df χ/df CFI RMSEA SRMR Loadings
Enjoyment of humor
SHS .*  . . . . .–.
SHS-P .**  . . . . .–.
Laughter
SHS .  . . . . .–.
SHS-P .  . . . . .–.
Verbal humor
SHS .***  . . . . .–.
SHS-P .*  . . . . .–.
Everyday life
SHS .  . . . . .–.
SHS-P .**  . . . . .–.
SHS .***  . . . . .–.
SHS-P .  . . . . .–.
Humor under stress
SHS .*  . . . . .–.
SHS-P .  . . . . .–.
Sense of humor score
SHS .***  . . . . .–.
SHS-P .***  . . . . .–.
Note: N = 857. Everyday life = Finding humor in everyday life.*p < 0.05. **p < 0.01. ***p < 0.001.

3.1.2 Item analyses
Table 3 shows the psychometric properties of the 24 SHS and SHS-P items.
As shown in Table 3, the CITCs were higher than the median of all remaining
correlations with the SHS and SHS-P scales, except for Item 19 of the SHS. This
Table 3: Psychometric properties of the 24 items of the Sense of Humor Scale (SHS) and the
Parallel Form of the SHS (SHS-P).
SHS items SHS-P items
Scale and item number M SD CITC Mdn r M SD CITC Mdn r
Enjoyment of humor
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
Laughter
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
Verbal humor
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
Everyday life
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
Humor under stress
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
 . . . . . . . .
Note: N = 857. Everyday life = Finding humor in everyday life, CITC = corrected item-total corre-
lations, Mdn r = median correlation with the five remaining scales of the SHS or SHS-P,
respectively.

item referred to the importance of having a large amount of humor in one’s life
and reflected SoH in general rather than its specific humor skill (enjoyment of
humor). CITCs ranged from 0.25–0.72 for the SHS items and from 0.43–0.80 for
the SHS-P items, underscoring that they were related to their corresponding
scale, but were not redundant. The median of the CITCs was numerically higher
for the SHS-P (0.66) than for the SHS items (0.58), and the discrimination to the
other scales was slightly better for the SHS-P (median correlation = 0.43) than for
the SHS items (0.48). All items showed sufficient variability (SD > 1) and were
approximately normally distributed. The means across the 24 SHS-P items were
numerically higher (M = 4.96, SD = 1.39) than those of the SHS (M = 4.54,
SD = 1.50). Both means were above the neutral answer option, indicating that
participants – on average – tended to agree with the items.
3.2 Parallel-test reliability

Next, parallel-test reliability is investigated, indicating the extent to which the
six SHS and SHS-P humor skills overlap. Table 4 shows the intercorrelations of
the SHS and SHS-P scales and the parallel-test reliabilities.
As shown in Table 4, parallel-test reliabilities (median 0.77) were high (with
the only values below 0.70 for enjoyment of humor and laughter) and always
numerically higher than the correlations with the other humor skills (median
0.51). The convergence of the SHS and SHS-P scales was also supported in CFA
models. The four items of each SHS and SHS-P scale were modeled to load on
one latent factor, and these two latent factors were allowed to correlate with one
another. Separate models were estimated for each of the six humor skills. The
factor correlations (which reflect the true-score correlations between the con-
structs, accounting for measurement error) ranged from 0.93 (laughter) to 1.00
(humor under stress) with a median of 0.97, indicating that the SHS and the
SHS-P humor skills virtually measured the same. Thus, the six scales of the SHS-
P can be regarded as sufficiently parallel to the SHS scales.
3.3 Psychometric properties of the combined SHS
Given the sufficient psychometric properties of the SHS-P and its high conver-
gence with the SHS, we aimed at combining their items to create more internally
consistent and factorially valid scales of the six humor skills. For this purpose,
Item 19 (SHS enjoyment of humor) was excluded, as it would have undermined
the separation of the six skills (see Table 3). Table 5 shows the descriptive

Table 4: Scale intercorrelations and parallel-test reliabilities of the Sense of Humor Scale (SHS) and the Parallel Form of the SHS (SHS-P).
Scales () () () () () () () () () () () () ()
Enjoyment of humor
() SHS
() SHS-P .
Laughter
() SHS . .
() SHS-P . . .
Verbal humor
() SHS . . . .
() SHS-P . . . . .
Everyday life
() SHS . . . . . .
() SHS-P . . . . . . .
() SHS . . . . . . . .
() SHS-P . . . . . . . . .
Humor under stress
() SHS . . . . . . . . . .
() SHS-P . . . . . . . . . . .
Total sense of humor
() SHS . . . . . . . . . . . .
Sense of Humor Scale
SHS-P . . . . . . . . . . . . .
Note: N = 857. Everyday life = finding humor in everyday life. Parallel-test reliabilities marked in italics.All correlations p < 0.001.

247
Table 5: Descriptive statistics, CITC ranges, internal consistencies, and scale intercorrelations
of the combined sense of humor scale.
Combined Sense M SD CITC range () () () () () () ()
of Humor
Scales
() Enjoyment of . . .–. .

humor
() Laughter . . .–. . .
() Verbal humor . . .–. . . .
() Finding humor . . .–. . . . .
in everyday life
() Laughing at . . .–. . . . . .
yourself
() Humor under . . .–. . . . . . .
stress
() Total sense of . . .–. . . . . . . .
humor
Note: N = 857. CITC = corrected item-total correlation. Cronbach’s alpha in italics.All correlations
p < 0.001.
statistics, CITC ranges, internal consistencies, and intercorrelations of the com-

bined SHS scales.
As shown in Table 5, the scales correlated positively and significantly with
one another (median = 0.55). Importantly, the internal consistencies of the scales
were always larger (0.76–0.93, median = 0.87), indicating that each scale had
unique reliable variance not shared with the other scales.
To test the factor structure of the combined SHS scales and to compare
different models with one another, the items were modeled as indicators of the
six humor skills in a CFA. Four different structural models were tested (see
Figure 1 for a depiction of each model): (a) A one-factor model (modeling the
SoH score only), (b) a six-factor model (modeling the six intercorrelated humor
skills), (c) a six-factor model with a second-order factor (modeling the six humor
skills and the SoH score in a hierarchical structure), and (d) a bifactor six-factor
model (modeling the six humor skills as correlated specific factors and the SoH
score as an independent general factor). In this model, each item can relate to
both SoH (i. e., a general sense of humor factor) and its specific humor skill (i. e.,
a specific group factor).
Table 6 shows the fit indices of these four CFA models based on the
combined SHS.

I 01
Enjoyment
…
of humor
I 19
I 02
Laughter
…
Item 01 I 20
Item 02 I 03
Verbal
…
Item 03 humor
Sense of I 21 Sense of
…
humor I 04 humor
Everyday
…
Item 45
I 22 life
Item 46
I 05
Item 47 Laughing
… at yourself
I 23
I 06
Humor un-
…
I 24 der stress
(a) One-factor model (c) Six-factor model with a second-order factor
I 01 I 01
Enjoyment Enjoyment
…
…
of humor of humor
I 19 I 19
I 02 I 02
Laughter Laughter
…
I 20 I 20
I 03 I 03
Verbal Verbal
…
humor humor
I 21 Sense of I 21
I 04 humor I 04
Everyday Everyday
…
I 22 life life
I 22
I 05 I 05
Laughing Laughing
…
I 23 at yourself at yourself
I 23
I 06 I 06
Humor un- Humor un-
…
I 24 der stress der stress

I 24
(b) Six-factor model (d) Bifactor six-factor model
Figure 1: Depiction of the four structural models of the combined Sense of Humor Scale tested
in the CFA. Everyday life = finding humor in everyday life.
As shown in Table 6, the one-factor model had an unacceptable fit, indicating

that more than one SoH factor is needed to explain the relationships between
the items. The bifactor six-factor model had the best fit, indicated by the lowest
AIC and BIC values and the best fit indices of all models. This supports the
importance of both a general SoH factor and each of the six humor skills. The
specific factors (i. e., humor skills) correlated between −0.06 (enjoyment of

Table 6: Fit indices of the CFA models of the humor skills of the combined Sense of Humor
Scale.
χ df χ/df CFI RMSEA SRMR AIC BIC
One-factor ʹ.***  . . . . ʹ. ʹ.

model
Six-factor model ʹ.***  . . . . ʹ. ʹ.
Hierarchical six- ʹ.***  . . . . ʹ. ʹ.
factor model
Bifactor six- ʹ.***  . . . . ʹ. ʹ.
factor model
Note: N = 857.***p < 0.001.
humor and verbal humor) and 0.62 (finding humor in everyday life and humor
under stress; median = 0.41). Despite measurement error being accounted for,
this median is numerically lower than the median of the observed intercorrela-
tions of the six humor skills, indicating that the general SoH factor contributed
to the relationships between the humor skills.
3.4 Overlap with other humor measures and life satisfaction

Having established the combined SHS scales as reliable and factorially valid,
their usefulness can be tested by allocating them in humorous temperament and
the four humor factors. Table 7 shows the correlations and multiple correlations
(derived from standard multiple regressions) of the six scales of the combined
SHS with humor-related attitude and mood, the STCI-T, the 4DHS, and the
SWLS.
As shown in Table 7 and in accord with McGhee’s assumptions, playful vs.
serious attitude and positive vs. negative mood correlated positively with all
combined SHS scales (small to large effects, median = 0.60 for attitude and 0.33
for mood). In addition, the six combined SHS scales explained a large amount of
the reliable variance in attitude and mood. The three temperamental bases of the
sense of humor explained between 14 % (enjoyment of humor) and 61 % (laugh-
ter) of the variance in the combined SHS scales (median 46 %). Also, each SHS
scale correlated significantly and positively with cheerfulness and negatively
with seriousness and bad mood (though the latter correlation was not significant
for enjoyment of humor). As expected, the combined SHS scales explained most
of the reliable variance in cheerfulness (suggesting that they were highly

Table 7: Zero-order correlations and multiple correlations (derived from standard multiple regressions) of the six scales and the Sense of Humor
(SoH) score of the combined Sense of Humor Scale with the State-Trait-Cheerfulness Inventory (STCI-T), the Four- Dimensions of Humor Scale (4DHS),
and the Satisfaction with Life Scale (SWLS).
Enjoy Laughter Verbal Everyday LAY Stress R R SoH score
Playful vs. serious attitude .*** .*** .*** .*** .*** .*** .*** . .***
Positive vs. negative mood .* .*** .*** .*** .*** .*** .*** . .***
STCI-T
Cheerfulness .*** .*** .*** .*** .*** .*** .*** . .***
Seriousness −.** −.*** −.** −.** −.*** −.*** .* . −.***
Bad mood −. −.** −.*** −.*** −.*** −.*** .*** . −.***
R .*** .*** .*** .*** .*** .*** .***
R . . . . . . .
DHS
Social fun .** .*** .*** .*** .*** .*** .*** . .***
Mockery . . .*** .* . . .* . .*
Humor ineptness .* −.* −. −.* −.*** −.** .*** . −.*
Cognitive/reflective .** .** .** .** .** .** .*** . .**
R .*** .*** .*** .*** .*** .*** .***
R . . . . . . .
SWLS .* .*** .*** .*** .*** .*** .*** . .***
Sense of Humor Scale
Note: N = 857 (attitude and mood), n = 177 (STCI-T), n = 140 (4DHS), and n = 513 (SWLS). Enjoy = enjoyment of humor, Verbal = verbal humor,
Everyday = finding humor in everyday life, LAY = laughing at yourself, Stress = humor under stress.*p < 0.05. **p < 0.01. ***p < 0.001.

251
similar). Overlaps with bad mood were also large, while they were somewhat
smaller with seriousness.
The 4DHS explained 13 % (enjoyment of humor) to 62 % (verbal humor) of
the variance in the combined SHS scales (median = 32 %). All combined SHS
scales correlated significantly and positively with social fun and cognitive/
reflective humor. Verbal humor, finding humor in everyday life, and the SoH
score also correlated positively with mockery, although these relationships were
numerically smaller than the correlations with social fun and cognitive/reflec-
tive humor. Most combined SHS scales also showed significant negative correla-
tions to humor ineptness, while verbal humor was uncorrelated with it and
enjoyment of humor correlated positively with it. The six combined SHS scales
overlapped most strongly with social fun, cognitive/reflective humor, and
humor ineptness, and to a lesser extent with mockery.
Lastly and as expected, the relationships with life satisfaction were positive
and significant for all combined SHS scales (median = 0.26). Still the size of
relationships varied across the humor skills, ranging from 0.10 (enjoyment of
humor) to 0.32 (laughter). Comparing the correlation of the SoH score and the
multiple correlations across the six humor skills across all outcomes, it can be
seen that the overlap was always numerically larger for the six skills than the
SoH score. This further supports the importance of taking the individual skills
into account.
4 Discussion
The first aim of the present study was to examine the psychometric properties of
the revised SHS (McGhee 1999). The internal consistencies of all of the skills and
the total score were > 0.70 with the exception of enjoyment of humor (0.56) and
laughter (0.65). This is in accord with previous findings (Proyer et al. 2010).
Unidimensionality (or homogeneity), as tested in the CFA, was supported for
four humor skills (enjoyment of humor, laughter, finding humor in everyday life,
and humor under stress), but not the SoH score. This implies that it is worth-
while to study the SHS humor skills in more detail, as they cannot be mapped
onto a single SoH factor. At the item level, the CITCs were sufficient and larger
than the correlations with the other humor skills. The only exception was Item
19 (“It is important for me to have a lot of humor in my life”), which was less
representative of the humor skill it was assigned to (enjoyment of humor), but
was related to SoH in general. Overall, the SHS scales had mixed psychometric
properties, with the SoH score being internally consistent (though not

unidimensional). The humor skills could, in general, be distinguished from one

another, yet their reliability could be improved. The revised SHS can thus be
recommended as a broad indicator of the sense of humor (by using the SoH
score across the six humor skills) and for reliably assessing the two humor skills
finding humor in everyday life and humor under stress.
The second aim was to develop a parallel form of the SHS (SHS-P) with four
items per humor skill. All skills and the total score of the SHS-P turned out to be
internally consistent (Cronbach’s alphas > 0.70). The skills were also unidimen-
sional while the total score was not; this finding further supports the idea that
the humor skills had a unique variance that cannot be explained by a single
sense of humor factor. Item CITCs were sufficient and they were always larger
than the correlations with the other five humor skills. Thus, the SHS-P was at
least as psychometrically sound as the SHS and, in some areas, superior. The
parallelism of the humor skills of the SHS and the SHS-P was supported in
sufficient parallel-test reliabilities (0.76–0.89). The lower values obtained for
enjoyment of humor (0.64) and laughter (0.67) might be due to the low internal
consistencies of these SHS scales. This interpretation is supported by the finding
that the latent correlations between the SHS and SHS-P factors in the CFA
ranged from 0.93 to 1.00, indicating that they were virtually identical once
measurement error was accounted for. This high degree of convergence con-
firmed the parallelism of the SHS and its parallel form, the SHS-P.
The third aim was to combine the items of the SHS and the SHS-P (to form
the combined SHS) to allow for a longer and therefore more internally consistent
and factorially valid measurement of the six humor skills. As Item 19 of the SHS
did not match with its humor skill (but with sense of humor in general), it was
excluded from the combined SHS as it would have hampered the separation of
the six humor skills. As expected, internal consistencies were good for all scales
(0.76 to 0.96). Comparing four different CFA models (one-factor, six-factor,
hierarchical six-factor, and bifactor six-factor model) yielded the best fit for
the bifactor six-factor model, in which each item was explained by a general
factor (presumably SoH) and one of the six humor skills.
The bifactor six-factor model model supported the notion that each item in the
combined SHS was determined by a general sense of humor component, which
was shared across all items, and the specific humor skill. The six humor skills
could be well distinguished in this model (median intercorrelation 0.41 vs. 0.55 in
the observed correlations). Also, their relationships might be captured more
realistically in this model than in the observed correlations, as the overlap
between the six humor skills was partly caused by the general sense of humor
factor that their items shared. For example, the observed correlation between
enjoyment of humor and verbal humor dropped from 0.38 to −0.06 in the bifactor

six-factor model, indicating that once general sense of humor was accounted for,
the two humor skills were essentially unrelated to one another. Passively seeking
out humorous stimuli and actively producing humor can thus be regarded as
independent skills. By contrast, the correlation between finding humor in every-
day life and humor under stress dropped only slightly from 0.75 to 0.62. Thus
these two humor skills were similar beyond general sense of humor, and they also
make use of similar skills (e. g., humor under stress could be conceived as finding
humor in stressful situations, representing a more elaborate and specific humor
skill similar to finding humor in everyday life).
The fourth aim was to investigate the overlap of the six humor skills with
humor-related measures and life satisfaction. Each humor skill and the SoH
score correlated significantly and positively with playful vs. serious attitude and
positive vs. negative mood, supporting McGhee’s (1999) idea that they represent
the “motor” of the sense of humor. The six combined SHS humor skills showed
an overlap of 74 % with cheerfulness, which supports the notion that they are
highly related and almost interchangeable concepts (as proposed by Ruch and
Carrell 1998). In addition, each SHS humor skill could be differentially located
within other humor-related concepts: They correlated positively and signifi-
cantly with social fun and cognitive/reflective humor, and negatively with
seriousness. All humor skills except enjoyment of humor correlated negatively
with bad mood. Verbal humor and finding humor in everyday life correlated
positively with mockery. Finally, enjoyment of humor correlated positively with
humor ineptness, verbal humor was uncorrelated with it, and the other four
humor skills correlated negatively with it. Thus, each humor skills covered
slightly different aspects of the sense of humor dimensions, and also the sizes
of their relationships varied. For example, laughing at yourself correlated more
strongly with humor ineptness than humor under stress did (−0.40 and −0.22,
respectively). Lastly, every humor skill and the SoH score were found to corre-
late positively with life satisfaction, with the smallest effects emerging for
enjoyment of humor and verbal humor and the strongest ones occurring for
laughter and laughing at yourself.
Overall, both the pattern of correlations across the six humor skills and the
size of the relationships varied. Similar effects can be expected for other mea-
sures of humor and well-being, which should be investigated in future research.
Importantly, assessing the individual humor skills allows evaluating the effec-
tiveness of practicing each humor skill in McGhee’s humor training program. For
example, it could show which of them (and their improvements by the humor
training) are more or less relevant for obtaining well-being outcomes. The
current study merely describes the existence of a relationship; causality needs
to be established in an experimental design.

4.1 Limitations and suggestions for future research
Both samples were convenience samples, in which females and well-educated

participants were overrepresented. Replications of the present findings (especially
regarding the performance of the combined SHS) with more varied samples are
desirable. The SHS was developed in the U.S., whereas the SHS-P was developed
in (German-speaking) Switzerland. Thus, replications with English-speaking and
more culturally diverse samples are needed. Regarding the factor structure of the
combined SHS, none of the CFA models had acceptable values in all fit indices.
The suggestion for the bifactor six-factor model should thus be seen as a pre-
liminary one. Specifically, optimizations of the combined SHS (i. e., by selecting
the best items for each humor skill) would be an important next step in future
research. This would potentially increase the fit in the CFA models and would
allow assessing the six humor skills more economically. Finally, other assessment
methods are needed to further support the validity of the (combined) SHS scales
(e. g., expert ratings, self-other-convergence, and behavioral observations).
4.2 Conclusions
In conclusion, the (revised) SHS can mainly be recommended if a general SoH

score (similar to cheerfulness) is of interest. The SoH score mainly spanned two
humor factors (social fun and cognitive/reflective humor), representing a spe-
cific subfacet of prosocial, entertaining, but at the same time thoughtful and
witty humor. This also fits well with Craik et al.’s (1996) finding that the lay
concept of the sense of humor is based on the socially warm (vs. cold) and
competent (vs. inept) styles of humorous conduct. The present study shifts the
focus from the SoH score to the six humor skills, which could be differentially
allocated in the four dimensions of the sense of humor. The SHS-P can be used
interchangeably with the SHS and additionally allows a reliable measurement of
each humor skill. The combined SHS can be recommended if the measurement
of the six humor skills is of primary interest. Being able to assess the individual
humor skills and by modeling them in the bifactor six-factor model can yield
insights into subfacets of the sense of humor that have not been frequently
studied so far. This can benefit both theory and practice by better understanding
the active ingredients in McGhee’s humor training program and the differential
relationships of the six humor skills to well-being and health.
Acknowledgements: We would like to thank Sarah Auerbach and the graduate

students for their help in creating the SHS-P items and in collecting data for

Sample 1. We would also like to thank Frank Appletree Rodden for his helpful
comments on an earlier version of this manuscript and Moritz Meyer for drawing
Figure 1.
References
Beermann, Ursula & Willibald Ruch. 2011. How virtuous is humor? What we can learn from
current instruments. The Journal of Positive Psychology 4. 528–539.
Craik, Kenneth, Martin D. Lampert & Arvalea Nelson 1996. Sense of humor and styles of
everyday humorous conduct. Humor: International Journal of Humor Research 9. 273–302.
Diener, Ed, Robert A. Emmons, Randy J. Larsen & Sharon Griffin 1985. The satisfaction with life
scale. Journal of Personality Assessment 49(1). 71–75.
McGhee, Paul E. 1979. Humor: Its origin and development. San Francisco: WH Freeman and
Company.
McGhee, Paul E. 1996. Health, healing and the amuse system: Humor as survival training.
Dubuque: Kendall/Hunt.
McGhee, Paul E. 1999. Health, healing and the amuse system: Humor as survival training, 3rd
edn. Dubuque: Kendall/Hunt.
McGhee, Paul E. 2010. Humor as survival training for a stressed-out world: The 7 humor habits
program. Bloomington: AuthorHouse.
Müller, Liliane & Willibald Ruch. 2011. Humor and strengths of character. The Journal of Positive
Psychology 6(5). 368–376.
Proyer, Rene T. 2014. A psycho-linguistic approach for studying adult playfulness: A replication
and extension toward relations with humor. The Journal of Psychology 148. 717–735.
Proyer, Rene T., Rahel Flisch, Stefanie Tschupp, Tracey Platt & Willibald Ruch. 2012. How does
psychopathy relate to humor and laughter? Dispositions toward ridicule and being
laughed at, the sense of humor, and psychopathic personality traits. International Journal
of Law and Psychiatry 35. 263–268.
Proyer, Rene T., Willibald Ruch & Liliane MüLler. 2010. Sense of humor among the elderly:
Findings with the German version of the SHS. Zeitschrift für Gerontologie und Geriatrie 43.
19–24.
R Core Team. 2015. R: A language and environment for statistical computing. Vienna:
R Foundation for Statistical Computing.
Rosseel, Yves. 2012. lavaan: An R package for structural equation modeling. Journal of
Statistical Software 48(2). 1–36.
Ruch, Willibald. 2012a. Towards a new structural model of the sense of humor: Preliminary
findings. AAAI Fall Symposium: Artificial Intelligence of Humor 2. 68–75.
Ruch, Willibald. 2012b. Four dimensions of humor scale. University of Zurich Unpublished
manual.
Ruch, Willibald & Amy Carrell. 1998. Trait cheerfulness and the sense of humor. Personality and
Individual Differences 24. 551–558.
Ruch, Willibald, Gabriele Köhler & Christoph van Thriel. 1996. Assessing the “humorous
temperament”: Construction of the facet and standard trait forms of the State-Trait-
Cheerfulness-Inventory – STCI. Humor: International Journal of Humor Research 9.
303–339.

Ruch, Willibald & Sonja Heintz. 2016. The virtue gap in humor: Exploring benevolent and
corrective humor. Translational Issues in Psychological Science 2. 35–45.
Ruch, Willibald & Paul E. McGhee. 2014. Humor intervention programs. In Acacia C. Parks &
Stephen M. Schueller (eds.), The Wiley Blackwell handbook of positive psychological
interventions, 179–193. Chichester: John Wiley & Sons.
Ruch, Willibald, Rene T. Proyer, Claudia Esser & Otilia Mitrache. 2011. Cheerfulness and
everyday humorous conduct. In Romanian Academy (eds.), Yearbook of the ‘George Baritiu’
Institute of History in Cluj-Napoca, 67–87. Bucharest: Romanian Academy Publishing
House.
Schermelleh-Engel, Karin, Helfried Moosbrugger & Hans Müller. 2003. Evaluating the fit of
structural equation models: Tests of significance and descriptive goodness-of-fit mea-
sures. Methods of Psychological Research Online 8(2). 23–74.
Soury, Mariette & Laurence Devillers. 2014. Smile and laughter in human-machine interaction:
A study of engagement. In Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn
Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk & Stelios Piperidis
(eds.), Proceedings of the ninth international conference on Language Resources and
Evaluation (LREC’14), 3633–3637. Paris: European Language Resources Association.
Wrench, Jason S. & James C. McCroskey 2001. A temperamental understanding of humor
communication and exhilaratability. Communication Quarterly 49(2). 142–159.
Bionotes
Willibald Ruch
Willibald Ruch is a Full Professor of Psychology at the University of Zurich, Switzerland. His
research interests are in the field of personality and assessment, with a special focus on humor
and laughter, cheerfulness, and smiling. In his doctoral dissertation at the University of Graz
(Austria) in 1980, he developed a taxonomy of jokes and cartoons and studied their relation to
personality. His more recent work, together with his research team at the University of Zurich,
includes humor from a positive psychology perspective, the effectiveness of humor training
programs and clown interventions, the ability to laugh at oneself, the fear of being laughed at
(gelotophobia), and the measurement of humor.
Sonja Heintz
Sonja Heintz is a postdoctoral researcher at the Department of Psychology at the University of

Zurich in Switzerland. Her main research interests in humor are individual differences (humor
and comic styles, dimensions of the sense of humor), measurement (humor questionnaires and
humor-related behaviors), and positive psychology (relationships of humor with character
strengths and well-being, virtuous forms of humor).


Multidimensional Sense of Humor Scale

Uploaded by

Copyright:

Available Formats

Multidimensional Sense of Humor Scale

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Multidimensional Sense of Humor Scale

Uploaded by

Copyright:

Available Formats

Zurich Open Repository and

Ruch, Willibald ; Heintz, Sonja

Posted at the Zurich Open Repository and Archive, University of Zurich

Originally published at:

Willibald Ruch* and Sonja Heintz

Keywords: Sense of Humor Scale, psychometrics, sense of humor model, humor

*Corresponding author: Willibald Ruch, Department of Psychology, University of Zurich, Zurich,

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

1.1 Aims of the present study

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

from 18 to 76 years provided valid responses. Subsets of this sample also

2.2.1 Sense of Humor Scale (SHS; as revised by McGhee 1999)

2.2.2 Parallel form of the Sense of Humor Scale (SHS-P)

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

2.2.3 Humor-related attitude and mood (McGhee 1999)

2.2.4 State-Trait-Cheerfulness-Inventory, Trait Version<60> (STCI-T<60>; Ruch

2.2.5 Four Dimensions of Humor Scale (4DHS; Ruch 2012b)

2.2.6 Satisfaction With Life Scale (SWLS; Diener et al. 1985)

The SWLS assesses general satisfaction with life, employing a seven-point

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

Sample 1 was recruited by several graduate students via personal invitations

2.4 Data analysis

Cronbach’s alpha and the mean inter-item correlations served as indicators of

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

3.1 Psychometric properties of the SHS and SHS-P scales

3.1.1 Scale-level analyses

Scales Ma SD Sk K Alpha MIIC rgender rage

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

of 0.80. The scales of the SHS-P showed higher internal consistencies,

Scales χ df χ/df CFI RMSEA SRMR Loadings

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

3.1.2 Item analyses

SHS items SHS-P items

Scale and item number M SD CITC Mdn r M SD CITC Mdn r

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

3.2 Parallel-test reliability

3.3 Psychometric properties of the combined SHS

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

Download Date | 5/2/18 5:07 PM

() Enjoyment of . . .–. .

statistics, CITC ranges, internal consistencies, and intercorrelations of the com-

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

(a) One-factor model (c) Six-factor model with a second-order factor

I 24 der stress der stress

(b) Six-factor model (d) Bifactor six-factor model

As shown in Table 6, the one-factor model had an unacceptable fit, indicating

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

χ df χ/df CFI RMSEA SRMR AIC BIC

One-factor ʹ.***  . . . . ʹ. ʹ.

Note: N = 857.***p < 0.001.

3.4 Overlap with other humor measures and life satisfaction

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

Enjoy Laughter Verbal Everyday LAY Stress R R SoH score

Download Date | 5/2/18 5:07 PM

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich

unidimensional). The humor skills could, in general, be distinguished from one

Brought to you by | UZH Hauptbibliothek / Zentralbibliothek Zürich