See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/269519251

Assessing social skills: the factorial structure and other

psychometric properties of four self-report measures

Article  in  Behavioral Psychology/Psicologia Conductual · December 2014

CITATION READS

1 5,371

5 authors, including:

Vicente E Caballo Isabel Salazar

University of Granada University of Granada
207 PUBLICATIONS   2,847 CITATIONS    117 PUBLICATIONS   1,677 CITATIONS   

SEE PROFILE SEE PROFILE

Pablo J. Olivares-Olivares Jose Olivares

University of Murcia Departamento Universitario Obrero Campesino
69 PUBLICATIONS   492 CITATIONS    115 PUBLICATIONS   4,577 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Cross-cultural assessment of social skills View project

Assessment and treatment of personality disorders View project

All content following this page was uploaded by Pablo J. Olivares-Olivares on 11 March 2019.

The user has requested enhancement of the downloaded file.

Behavioral Psychology / Psicología Conductual, Vol. 22, Nº 3, 2014, pp. 375-399

ASSESSING SOCIAL SKILLS: THE FACTORIAL STRUCTURE AND

OTHER PSYCHOMETRIC PROPERTIES OF FOUR SELF-REPORT
MEASURES

Vicente E. Caballo1, Isabel C. Salazar2, Pablo Olivares3, María Jesús

Irurtia4, José Olivares3, and Rafael Toledo5
1
University of Granada; 2FUNVECA Clinical Psychology Center, Granada; 3University
of Murcia; 4University of Valladolid (Spain); 5University of North Texas (USA)

Abstract
The self-reporting of social skills continues to be assessed through the
assertiveness inventories developed in the 1970s, such as the Rathus
Assertiveness Schedule (RAS), the Assertion Inventory (AI), and the College Self-
Expression Scale (CSES). The study reported here involved 421 university students
(76.5% women) and obtained the factor structures of the aforementioned
instruments, plus the new Social Skills Questionnaire (SSQ-I) (Cuestionario de
habilidades sociales, CHASO-I). The factorial solutions obtained were 6, 8, 11 and
12 factors, respectively. The reliability (Guttman split-half and Cronbach’s alpha)
of all the questionnaires was high, and the correlations between the CHASO-I
and all the other questionnaires were moderate. The sex differences found
involved the total scores of the RAS and the factors “Speaking or performing in
public/Interacting with figures in authority”, “Interacting with persons I am
attracted to”, and “Interacting with strangers”, with men being more skilled than
women, and the factor “Apologizing/Recognizing their own mistakes”, with
women being more skilled than men. The study concluded by recognizing certain
common problems affecting the self-report measures of social skills, as well as
certain advantages of the new CHASO-I.
KEY WORDS: social skills, factor structure, questionnaires, assertiveness, sex
differences.

Resumen
La evaluación de las habilidades sociales por medio de medidas de
autoinforme sigue realizándose con instrumentos desarrollados en los años 70,
como el “Inventario de asertividad de Rathus” (RAS), el “Inventario de aserción”
(AI) o la “Escala de autoexpresión universitaria” (CSES). Este estudio llevado a
cabo con 421 estudiantes universitarios (76,5% mujeres) halló las estructuras
factoriales de dichos instrumentos así como del nuevo “Cuestionario de


This study was partially supported by a grant from the European Regional Development Fund (ERDF)
and Junta de Andalucía as part of the “Fortalecimiento de la I+D+i” program of the University of
Granada.
Correspondence: Vicente E. Caballo, Faculty of Psychology, University of Granada, 18071 Granada
(Spain). E-mail: [email protected]
376 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

habilidades sociales” (CHASO-I). Las soluciones factoriales obtenidas fueron de 6,

8, 11 y 12 factores respectivamente. La fiabilidad (dos mitades de Guttman y
alfa de Cronbach) de todos los cuestionarios fue alta y las correlaciones entre el
CHASO-I y el resto de los cuestionarios fueron moderadas. Se encontraron
diferencias de sexo en la puntuación total del RAS y en los factores de “Hablar o
actuar en público/Interaccionar con superiores”, “Interaccionar con personas que
me atraen” e “Interaccionar con desconocidos”, con los hombres más habilidosos
que las mujeres, y en el factor “Pedir disculpas/Reconocer errores propios”, con
las mujeres más habilidosas que los hombres. El estudio concluye señalando
algunos problemas comunes a las medidas de autoinforme de las habilidades
sociales así como algunas virtudes del nuevo CHASO-I.
PALABRAS CLAVE: habilidades sociales, estructura factorial, asertividad,
cuestionarios, diferencias de sexo.

Introduction

Interest in the assessment of social skills peaked in the 1970s and 1980s
(usually under the name of “assertiveness” or “assertion” assessment), and today
we still profit from the major research conducted in those years. Furthermore, the
assessment of social skills is, still today as it was then (see Caballo, 1986, 1997), a
controversial issue pending general consensus among experts in the field. Like
then, too, we may refer to the problem with a meaningful title reflecting its
difficulty: “Pandora's Box reopened? The assessment of social skills” (Curran,
1979).
If we focus on self-report measures assessing social skills, we find that most
of the questionnaires used today were developed in those years. In spite of the
time lapse, it is unusual to find relevant research describing the psychometric
properties of those measures, leaving behind a statistical “limbo”. In fact, as
opposed to other areas, we do not have a self-report assessment measure that
acts as a gold standard for comparing or validating other measures. This may be
due, at least partially, to the fact that many of the questionnaires were developed
with a clinical application in mind (e.g., assessing the effectiveness of assertiveness
or social skills training), or as an ad hoc tool for specific research, but without
considering a subsequent and wider application.
The fact is that we are currently facing several important issues in regard to
the psychometric nature of the social skills self-report measures. One such issue
involves factorial validity. There seems to be some consensus that when speaking
about social skills we are taking a series of response classes or behavioral
dimensions for granted (Caballo, 1997; Galassi & Galassi, 1977; Lazarus, 1973).
Social skills assessment questionnaires should include such dimensions. The
problem is that there is no consensus over their number and nature. As many as
14 possible different dimensions or response classes have been proposed (see
Caballo, 1997, for a review), but they are not included in any self-report
assessment measures, especially not as non-overlapping independent entities. An
analysis of the factorial structure of social skills questionnaires has found that the
number of identified factors varies according to the specific measure and the
 Assessing social skills with four self-report measures 377

particular study (somewhat akin to what happens in the field of social anxiety, as
reported by Caballo, Salazar, Irurtia, Arias, & Nobre [2013], although it is much
more evident in the social skills field).
There are no stable factorial structures using the same questionnaire. For
instance, in the case of the Rathus Assertiveness Schedule (RAS; Rathus, 1973),
there are very different factorial solutions, varying from four (Heimberg & Harrison,
1980), through seven (Carrasco, Clemente, & Llavona, 1983), eight and nine
(Nevid & Rathus, 1979), up to 12 factors (Hull & Hull, 1978), even though Rathus
published the scale as a general measure of assertiveness. Something similar
happens with the Adult Self-Expression Scale (ASES; Gay, Hollandsworth, Jr., &
Galassi, 1975), where the authors reported a structure of 14 factors, although
LaFromboise (1983) subsequently found several factorial solutions (14, 13, and,
finally, three) in an attempt to reduce the number of items. The same may be said
regarding the Personal Relations Inventory (PRI; Lorr & More, 1980), in which the
authors proposed a 4-factor solution, although different solutions were found in
further studies that included samples from other countries, such as the nine factors
in Australia (Heaven, 1984) or the five factors in Chile (Carmona & Lorr, 1992).
The Assertion Inventory (AI; Gambrill & Richey, 1975) also varies significantly
regarding the number of factors depending on the studies: 11 factors in the
former original study, 13 factors according to Carrasco, Clemente, and Llavona
(1989), and eight in the study by Castaños, Reyes, Rivera, and Díaz (2011). The
same thing happens with the College Self-Expression Scale (CSES; Galassi, DeLo,
Galassi, & Bastien, 1974), in which four (Kipper & Jaffe, 1978), six (Galassi &
Galassi, 1979), 11 (Caballo & Buela, 1988), and even 14 (Henderson & Furnham,
1983) factors have been found.
The inconsistency of the factorial structures of the self-report measures of
social skills may be due to a variety of reasons. One could be that too many factors
are often extracted and retained, in spite of having very few items (one or two), as
was the case, for example, with the AI. Another reason is that there are no
consistent standards for retaining specific items for each factor, that is, no attempt
is made to retain only those items with loadings above a particular score on a
factor (e.g., .35 or more) or, less frequently, to discard those items with high
loadings on two or more factors. Thus, in the specific case of the Wolpe-Lazarus
Assertiveness Scale (WLAS; Wolpe & Lazarus, 1966), while Henderson and
Furnham (1983) used .40 as the loading criterion to include an item in a factor,
Hersen et al. (1979) used .25, and Kogan et al. (1996) used .30. In the case of the
CSES, Kipper and Jaffe (1978) used .25, while Galassi and Galassi (1979) used .35
as the loading criterion.
A further problem with social skills questionnaires is that identical names are
often given to two separate factors in spite of their different composition. For
instance, Hersen et al. (1979) analyzed the factorial structure of the WLAS
separating males from females, and although the solution in both samples
revealed 10 factors, the first factor (which explained the highest variance
percentage in both samples), called “General Expressivity”, contained different
items. The items in the male sample were 12, 14, 16, 17, 28, and 29; and 5, 11,
16 and 28 in the female sample. This first factor shared only two items in these
378 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

samples, but was given the same name in both cases. Additionally, three out of
the six items that formed the factor in the males sample loaded also on another
factor. The situation seems to deteriorate when analyzing a combined male and
female sample with the same questionnaire (WLAS), as in the work by Henderson
and Furnham (1983), where that “same” factor has a very different composition
to the one found in the work by Hersen et al. (1979), including items 12, 15, 21
and 22. In sum, although both studies found the same factorial solution, the items
included in each factor do not match, and the variance explained by the main
factors differs significantly. This can also be seen with the AI in the research by
Gambrill and Richey (1975), and Henderson and Furnham (1983), in which the
solution was 11 factors in each one, but only three factors have a major
concordance, some concordance in four, and none at all in five of the factors.
Recently developed questionnaires share the same problems. For instance, in
the case of the Social Skills Inventory (Inventário de Habilidades Sociais, IHS; Del
Prette, Del Prete, & Barreto, 1998) the same factorial solution was proposed (five
factors) in two studies (Del Prette et al., 1998; Olaz, Medrano, Greco, & Del Prette,
2009), but there was no coincidence of the items in each factor. Only in one factor
did most items coincide, with partial coincidence in three of the factors (fewer
than half of the items), while the fifth factor is completely different in both studies.
Considering the problems mentioned in this review, we decided to analyze
the factorial structure of three assertiveness self-report measures that are widely
used around the world today, namely, the RAS, the AI, and the CSES, as well as
the Social Skills Questionnaire (SSQ-I) (Cuestionario de habilidades sociales-I,
CHASO-I), a new measure recently developed by our team to evaluate social skills.

Method

Participants

The sample consisted of 421 participants with a mean age of 22.54 years
(SD= 6.75), being made up of 322 females (M= 22.25 years, SD= 6.01) and 99
males (M= 23.49, SD= 8.70). They were mostly university students (93.82%). Of
these, 246 were Psychology students, and 149 were students in other subjects. As
regards the rest of the sample (6.18%), one was a psychologist, seven were
workers with other university degrees, four were workers without university
studies, two were pre-university students, and eight were undefined (e.g., jobless,
retired, etc.) There were no data on four of the participants.
However, not all the subjects completed all the questionnaires. The entire
sample answered the Social Skills Questionnaire (SSQ-I) (Cuestionario de
habilidades sociales, CHASO-I) and the Rathus Assertiveness Schedule (RAS), but
the participants only partially completed the Assertion Inventory (AI) and the
College Self-Expression Scale (CSES). In fact, the AI was answered by 228 subjects
with a mean age of 21.62 years (SD= 7.00), of whom 170 were females (M=
21.26 years, SD= 5.91) and 50 were males (M= 22.67 years, SD= 9.48), while the
CSES was completed by 117 subjects, with a mean age of 21.53 years (SD= 6.82),
 Assessing social skills with four self-report measures 379

and consisted of 88 females (M= 21.08 years, SD= 5.27) and 29 males (M= 22.93
years, SD= 10.19).

Instruments

The following self-report instruments were used to assess assertiveness in this

study:
 Social Skills Questionnaire (SSQ-I) (“Cuestionario de habilidades sociales”,
CHASO-I; Caballo, Salazar, & Irurtia, 2014). This first version of the
questionnaire includes 116 items (plus two control items) that assess different
social skills dimensions selected both from the literature in the field (see
Caballo, 1997) and from the factors composing the Multidimensional Social
Expression Scale – Motor Part (Escala multidimensional de la expresión social -
parte motora, EMES-M; Caballo, 1993), and the Social Anxiety Questionnaire
for Adults (SAQ-A30; Caballo et al., 2010; Caballo, Salazar, et al., 2012). The
dimensions to be assessed are as follows: 1) Interaction with people I am
attracted to [11 items], 2) Standing up for my legitimate rights [12 items], 3)
Speaking in public/Interaction with authority figures [9 items], 4) Public
performance [6 items], 5) Acknowledging ignorance or a mistake and
apologizing [14 items], 6) Interacting with strangers [17 items], 7) Giving
compliments/Expressing positive feelings [13 items], 8) Accepting compliments
and positive feelings [6 items], 9) Asking for favors/Making requests [11 items],
10) Rejecting requests [8 items], and 11) Expressing personal opinions [9
items]. Each item is answered on a 5-point Likert scale from 1 (“Very
uncharacteristic of me”) to 5 (“Very characteristic of me”). No item was
negatively formulated, so the total score and the dimensions score are the sum
of all or part of the items, and the higher the score, the greater the social skill.
 Rathus Assertiveness Schedule (RAS; Rathus, 1973). This 30-item instrument
was designed to measure assertiveness. Each item is answered from +3 “Very
characteristic of me, extremely descriptive” to -3 “Very uncharacteristic of me,
extremely non-descriptive”, without including 0. There are 17 items that
require inverting the sign, and then the items’ scores are added up. A high
positive score means high social skills, while a high negative score means the
opposite. The test-retest reliability found was between .76 and .83, split-half
reliability was .77, and internal consistency (Cronbach’s alpha) ranged
between .73 and .86 (Beck & Heimberg, 1983; Heimberg & Harrison, 1980;
Rathus, 1973; Vaal, 1975). The RAS has been criticized by Galassi and Galassi
(1975) for not distinguishing between assertiveness and aggressiveness.
 Assertion Inventory (AI; Gambrill & Richey, 1975). The AI was developed to
gather three kinds of information regarding assertive behavior: a) degree of
discomfort or anxiety when handling a variety of social situations, b) response
probability of engaging in assertive behavior in these same social situations,
and c) identification of situations in which the respondent would like to be
more assertive. This study considered only the second type of information
regarding assertive behavior (b). The inventory consists of 40 items, and each
item on the Response Probability subscale is answered on a 5-point Likert scale
380 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

from 1 (“Always do it”) to 5 (“Never do it”); the higher the score, the lower
the assertiveness. The authors of the questionnaire found a test-retest
reliability of .81 for the Response Probability subscale (Gambrill & Richey,
1975). Reports on this subscale in Spanish samples have found reasonable
test-retest reliability (r = .84) and internal consistency (Cronbach’s alpha = .88)
(Carrasco et al., 1989; Casas-Anguera et al., 2014).
 College Self-Expression Scale (CSES; Galassi, Delo, Galassi, & Bastien, 1974).
This scale contains 50 items using a 5-point Likert scale from 0 (“Almost
always or always”) to 4 (“Never or rarely”). Twenty-nine items are worded so
that they require reverse scoring. Scores on the 50 items are summed to yield
a total score, with higher total scores indicating more assertion. Its
psychometric properties have been considered as appropriate. The test-retest
reliability found varies between .89 and .90 (Galassi et al., 1974). The test-
retest reliability found with a Spanish university sample was .87, and the
internal consistency (Cronbach’s alpha) was .89 (Caballo & Buela, 1988).

Procedure

The questionnaires were collectively administered in classrooms at the

universities of Granada, Murcia and Valladolid, and in groups for non-students.
The questionnaires were answered anonymously to respect personal privacy. For
technical reasons, the number of questionnaires administered varied throughout
the study, so the time required to fill out the questionnaires also varied
accordingly.

Statistical analysis

An exploratory factor analysis (EFA) with a hierarchical analysis of oblique

factors was performed by means of Statistica v. 12.0 (StatSoft, 2013) for each one
of the questionnaires administered. The hypothesis behind this type of factor
analysis assumes a general factor (secondary) of social skill that probably affects all
the social skills dimensions assessed by each questionnaire. Furthermore, the
factorial solutions obtained with this kind of analysis are neater and clearer,
favoring the items loading on only one factor. An item was considered to load
significantly on a factor if its value was at least ±.40 on that factor for
interpretation purposes (Stevens, 1986).

Results

Social Skills Questionnaire (SSQ-I) (“Cuestionario de habilidades sociales”, CHASO-I)

The EFA identified 12 factors with eigenvalues higher than 1.00 as the best
factor solution for the SSQ-I, explaining 48.05% of the cumulative variance. Table
1 shows these 12 factors, each including two of the most representative items of
the corresponding factor fulfilling the following criteria: 1) a high loading on the
 Assessing social skills with four self-report measures 381

factor (and always higher than .40), and 2) helping to operationally define the
factor. Furthermore, the eigenvalues and explained variance of each factor are
included.

Table 1
Exploratory factor analysis of the Social Skills Questionnaire (SSQ-I) and loadings of two
representative items for each factor (N= 421)

Loadings of two SSQ-I Factors

representative items
for each factor of the F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11 F12 ri-t
SSQ-I
Factor 1. Speaking or performing in public/Interacting with people in authority (eigenvalue: 23.29;
explained variance: 20.08%)
35. Performing or
acting out a scene in
.69 .04 .04 .13 -.24 .13 -.01 .01 -.07 -.04 .17 -.00 .42
front of a group of
strangers
69. Participating in a
meeting with people .63 .05 .14 .19 .19 -.04 -.04 .10 .11 .00 .10 -.04 .47
in authority
Factor 2. Expressing positive feelings (eigenvalue: 7.19; explained variance: 6.20%)
56. Expressing
affection (kiss, hug, .03 .67 .05 .19 -.12 -.01 -.06 .03 -.01 .21 -.05 -.01 .39
caress) to a loved one
92. Expressing support
(hug, caress) to a close
-.02 .64 .05 .05 -.16 .10 -.01 .10 .13 .18 .09 .08 .43
person when she/he
needs it
Factor 3. Refusing requests (eigenvalue: 5.12; explained variance: 4.41%)
106. Refusing to do
something I do not .03 .12 .76 .05 -.09 -.05 .00 .10 .00 .03 .10 -.00 .35
want to do
111. Refusing what I
consider to be an .09 .10 .71 .04 .05 -.02 .04 .07 .15 .18 .16 -.04 .46
unreasonable request
Factor 4. Interacting with persons I am attracted to (eigenvalue: 3.33; explained variance: 2.87%)
13. Asking someone I
find attractive for a .10 .01 .00 .86 .02 .04 -.01 .08 .02 .01 .06 -.05 .43
date
30. Asking someone I
.12 .05 .04 .85 .01 .07 -.00 .01 .06 .05 .04 .04 .48
like for a date
Factor 5. Dancing or singing in public (eigenvalue: 2.70; explained variance: 2.33%)
19. Dancing in front of
-.16 -.19 -.06 -.10 .51 -.12 -.11 -.06 .05 -.06 -.33 -.06 .37
other people
71. Singing in public -.30 .07 .16 -.15 .41 -.17 .14 -.10 .06 -.12 -.06 .12 .21
Factor 6. Disclosing information about myself to close persons (eigenvalue: 2.55; explained
variance: 2.19%)
45. Talking about
personal matters with .00 .22 -.01 .13 -.09 .62 .16 .04 .18 .10 .16 .06 .48
a friend
382 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

54. Disclosing personal

.00 .20 .00 .15 -.14 .62 .07 .03 .21 .11 .12 .10 .49
information to a friend
Factor 7. Asking an attendant or a stranger for something (eigenvalue: 2.12; explained variance:
1.83%)
43. Asking an
.13 .14 .13 .01 -.05 .23 .50 .30 .01 .16 .20 -.11 .51
attendant for service
46. Asking someone
on the street for the
.12 .11 .02 .03 .08 .25 .40 .20 -.09 .32 .25 .06 .49
time, an address or a
shop
Factor 8. Expressing annoyance, disgust, or displeasure (eigenvalue: 2.08; explained variance:
1.80%)
24. Telling someone to
stop being annoying .13 .00 .21 .09 -.06 -.01 .16 .67 .05 -.02 .08 .07 .42
or making noise
12. Telling someone
.09 -.03 .15 .11 .02 -.05 .09 .61 .15 .08 .09 -.00 .40
not to jump the queue
Factor 9. Expressing different opinions/Clarifying opinions (eigenvalue: 2.00; explained variance:
1.72%)
87. Disagreeing with
others when I think I .07 .06 .27 .07 -.07 .21 -.03 .20 .53 .04 .11 .01 .40
am right
104. Clarifying my
opinion in a group
.25 .15 .25 -.02 -.07 .03 -.01 .16 .42 .22 .28 .02 .53
when I have not been
understood
Factor 10. Apologizing/Recognizing my own mistakes (eigenvalue: 1.96; explained variance:
1.69%)
2. Apologizing if my
behavior has upset .09 .18 -.01 .05 .01 -.01 -.04 -.03 -.02 .75 .03 -.04 .33
someone
20. Apologizing when
.12 .12 .07 .01 -.08 .02 .04 .00 .06 .74 .10 -.07 .45
I am wrong
Factor 11. Interacting with strangers (eigenvalue: 1.75; explained variance: 1.51%)
84. Talking to people I
do not know at parties .26 .04 .09 .21 -.01 .09 .07 .08 .15 .15 .67 -.01 .63
and gatherings
110. Behaving in a
extroverted way in .24 .09 .09 .20 -.05 .21 -.05 .14 .02 .06 .60 .04 .57
new social situations
Factor 12. Thanking for congratulations/Arguing with salespeople (eigenvalue: 1.65; explained
variance: 1.42%)
38. Thanking for well-
wishes received on my .07 .32 .07 -.01 .05 .07 .02 .08 -.05 .33 .00 .54 .31
birthday
82. Arguing about the
price with salespeople -.24 -.08 -.04 -.18 -.21 -.11 -.01 -.27 .14 .04 -.05 .43 .32
or attendants
Note: SSQ-I= Social Skills Questionnaire-I; ri-t= item-total correlations.

The 12 factors and the items loading on each one of them were the
following: Factor 1: Speaking or performing in public/Interacting with people in
 Assessing social skills with four self-report measures 383

authority (items 3, 26, 31, 35, 37, 49, 62, 67, 69, 78, 94, 102); Factor 2:
Expressing positive feelings (items 9, 18, 56, 64, 81, 83, 89, 92, 96, 99, 101, 105,
113); Factor 3: Refusing requests (items 8, 21, 65, 100, 106, 111); Factor 4:
Interacting with persons I am attracted to (items 5, 13, 22, 30, 36, 50, 79); Factor
5: Dancing or singing in public (items 19, 71); Factor 6: Disclosing information
about myself to close persons (items 45, 54, 107, 116); Factor 7: Asking an
attendant or a stranger for something (items 43, 46, 80); Factor 8: Expressing
annoyance, disgust, or displeasure (items 4, 12, 14, 23, 24, 32, 40, 86, 90); Factor
9: Expressing different opinions/Clarifying opinions (items 7, 75, 87, 91, 104);
Factor 10: Apologizing/Recognizing my own mistakes (items 2, 15, 17, 20, 48, 93,
115, 118); Factor 11: Interacting with strangers (items 16, 34, 47, 57, 84, 103,
110, 114); Factor 12: Thanking for congratulations/Arguing with salespeople
(items 38, 82). The remaining items did not load above .40 on any factor (except
for three items loading on two factors and remaining outside the factorial
solution).
The questionnaire’s internal consistency (Cronbach’s alpha) was .97 and the
split-half reliability (Guttman) was .96.

Rathus Assertiveness Schedule (RAS)

The EFA identified six factors with eigenvalues higher than 1.00 as the best
factor solution for the RAS, explaining 44.79% of the cumulative variance. Table 2
shows these six factors, each including two of the most representative items of the
corresponding factor fulfilling the same criteria as with the SSQ-I. Furthermore, the
eigenvalues and explained variance of each factor are included.

Table 2
Exploratory factor analysis of the Rathus Assertiveness Schedule (RAS; Rathus, 1973) and
loadings of two representative items for each factor (N= 421)

Loadings of two representative items RAS Factors

ri-t
for each factor of the RAS F1 F2 F3 F4 F5 F6
Factor 1. Interacting with commercial firms (eigenvalue: 5.24; explained variance: 17.47%)
12. I will hesitate to make phone calls
to business establishments and .60 .17 .11 .12 .07 -.04 .40
institutions
14. I find it embarrasing to return
.59 .28 .06 -.01 .01 .09 .40
merchandise
Factor 2. Expressing annoyance or displeasure (eigenvalue: 2.26; explained variance: 7.53%)
3. When the food served at a
restaurant is not done to my
.08 .80 -.07 .01 .04 .05 .35
satisfaction, I complain about it to the
waiter or waitress
25. I complain about poor service in a
-.03 .79 .02 .03 -.03 .00 .30
restaurant and elsewhere
Factor 3. Defending my own position (eigenvalue: 1.83; explained variance: 6.10%)
6. When I am asked to do something, I
.11 -.00 .71 -.03 -.04 .02 .14
insist upon knowing why
384 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

8. I strive to get ahead as well as most

.24 -.02 .60 .06 .07 .00 .22
people in my position
Factor 4. Interacting with others in an extroverted way (eigenvalue: 1.49; explained
variance: 4.97%)
2. I have hesitated to make or accept
.17 .08 .01 .69 .03 .12 .44
dates because of “shyness”
10. I enjoy starting conversations with
.03 -.02 .09 .53 .27 -.10 .27
new acquaintances and strangers
Factor 5. Expressing feelings openly (eigenvalue: 1.38; explained variance: 4.59%)
21. I am open and frank about my
.14 .03 -.05 .14 .75 -.11 .23
feelings
20. When I have done something
important or worthwhile, I manage to -.19 -.02 .15 -.21 .60 .01 -.06
let others know about it
Factor 6. Refusing requests (eigenvalue: 1.23; % explained variance: 4.11)
4. I am careful to avoid hurting other
people’s feelings, even when I feel that -.14 .11 .16 .14 -.26 .70 .28
I have been injured
23. I often have a hard time saying
.39 .14 .04 .07 .19 .53 .50
“No”
Note: RAS= Rathus Assertiveness Schedule; ri-t= item-total correlations

The six factors and the items loading on each one of them were the
following: Factor 1: Interacting with commercial firms (items 12, 13, 14, 16, 17);
Factor 2: Expressing annoyance or displeasure (items 3, 25, 27, 28). Factor 3:
Defending my own position (items 6, 7, 8, 22); Factor 4: Interacting with others in
an extroverted way (items 1, 2, 10, 11, 18, 30); Factor 5: Expressing feelings
openly (items 20, 21, 24, 26); Factor 6: Refusing requests (items 4, 5, 15, 23).
Items 9, 19 and 20 did not load above .40 on any factor, or did so on two or more
factors.
The internal consistency of the questionnaire (Cronbach’s alpha) was .82 and
the split-half reliability (Guttman) was .80.

Assertion Inventory (AI)

The EFA identified eight factors with eigenvalues higher than 1.00 as the best
factor solution for the AI (subscale of Response Probability), explaining 52.34% of
the cumulative variance. Table 3 shows these eight factors, each including two of
the most representative items of the corresponding factor fulfilling the same
criteria as with the SSQ-I. Furthermore, the eigenvalues and explained variance of
each factor are included.
The eight factors and the items loading on each one of them were the
following: Factor 1: Expressing annoyance, disgust or displeasure (items 13, 28,
35, 39, 40); Factor 2: Making requests to potential employers (items 9, 17, 18, 33);
Factor 3: Giving and receiving compliments, requests and asking personal
questions (items 2, 3, 12, 16, 21, 29, 30, 37); Factor 4: Resisting undesirable
propositions (items 10, 25, 27, 31, 32, 34); Factor 5: Resisting consumer
propositions (items 4, 6, 23); Factor 6: Refusing requests involving my own
 Assessing social skills with four self-report measures 385

property and asking for it to be returned (items 11, 36); Factor 7:

Apologizing/Recognizing my own mistakes (items 5, 7, 8, 14, 15); Factor 8:
Expressing different opinions/Clarifying opinions (items 19, 26, 38). Items 1, 20, 22
and 24 did not load above .40 on any factor, or did so on two or more factors.
The internal consistency of the assertiveness subscale (Cronbach’s alpha) was
.90 and the split-half reliability (Guttman) was .93.

Table 3
Exploratory factor analysis of the subscale “Response probability” of the Assertion Inventory
(AI; Gambrill & Richey, 1975) and loadings of two representative items for each factor (n=
228)

Loadings of two AI-RP Factors

representative items for each ri-t
F1 F2 F3 F4 F5 F6 F7 F8
factor of the AI-RP
Factor 1. Expressing annoyance, disgust or displeasure (eigenvalue: 8.63; explained variance:
21.59%)
39. Tell a friend or co-
worker when he or she
.83 .13 .07 .09 .06 .00 -.01 .09 .48
says or does something
that bothers you
40. Ask a person who is
annoying you in a public .77 -.04 .11 .04 .01 .04 -.05 -.03 .36
situation to stop
Factor 2. Making requests to potential employers (eigenvalue: 2.59; explained variance:
6.46%)
18. Your initial request for a
meeting is turned down and
.00 .78 .07 -.00 -.04 .02 .02 .02 .30
you ask the person again at
a later time
17. Request a meeting or a
-.03 .66 .39 -.01 -.07 .11 .06 .22 .49
date with a person
Factor 3. Giving and receiving compliments, requests and asking personal questions
(eigenvalue: 2.31; explained variance: 5.77%)
16. Compliment a person
you are romantically involved .04 -.01 .70 .20 -.04 -.05 .21 .02 .39
with or interested in
2. Compliment a friend .08 .03 .69 .21 -.00 .03 .20 .06 .47
Factor 4. Resisting undesirable propositions (eigenvalue: 1.98; explained variance: 4.96%)
34. Resist pressure to use
.01 .04 .09 .68 .03 -.00 .12 .12 .27
drugs
27. Resist sexual overtures
.20 -.10 .07 .65 .01 .36 .06 .03 .35
when you are not interested
Factor 5. Resisting consumer propositions (eigenvalue: 1.45; explained variance: 3.64%)
4. Resist sales pressure -.02 -.07 -.01 .15 .71 .03 .10 .12 .16
6. Turn down a request for a
.13 .41 -.07 .18 .52 .14 .02 -.29 .29
meeting or a date
386 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

Factor 6. Refusing requests involving my own property and asking for it to be returned
(eigenvalue: 1.38; explained variance: 3.45%)
11. Turn down a request to
.08 .27 .05 .06 .06 .70 -.07 -.03 .33
borrow money
1. Turn down a request to
-.04 -.12 .05 .02 .49 .53 -.00 .16 .24
borrow your car
Factor 7. Apologizing/Recognizing my own mistakes (eigenvalue: 1.32; explained variance:
3.29%)
5. Apologize when you are
-.14 .02 .22 .19 .17 -.12 .65 .16 .32
at fault
7. Admit fear and request
.19 .26 .33 .07 .14 .05 .47 -.24 .45
consideration
Factor 8. Expressing different opinions/Clarifying opinions (eigenvalue: 1.27; explained
variance: 3.17%)
38. Continue to converse
with someone who .20 .13 .19 .20 .15 -.01 .16 .55 .48
disagrees with you
26. Express an opinion that
differs from that of the
.32 -.01 .25 .27 .10 .06 .15 .51 .53
person with whom you are
talking
Note: AI-PR= Assertion Inventory-Response probability; ri-t= item-total correlations.

College Self-Expression Scale (CSES)

The EFA identified 11 factors with eigenvalues higher than 1.00 as the best
factor solution for the CSES, explaining 56.86% of the cumulative variance. Table
4 shows these 11 factors, with each one including two of the most representative
items of the corresponding factor fulfilling the same criteria as with the SSQ-I.
Furthermore, the eigenvalues and explained variance of each factor are included.
The 11 factors and the items loading on each one of them were the
following: Factor 1: Expressing annoyance, disgust, or displeasure (items 6, 9, 13,
18, 23, 28, 30, 32, 38, 41, 47, 48); Factor 2: Expressing positive feelings (items 10,
20, 36, 49); Factor 3. Speaking in public or in front of the class (items 21, 33, 43,
50); Factor 4: Defending personal preferences before my parents (items 29, 46);
Factor 5: Giving compliments to friends (items 25, 26, 31); Factor 6: Interacting
with the opposite sex (items 22, 39, 44); Factor 7: Defending my own rights before
friends/roommates (items 11, 16, 27, 34, 45); Factor 8: Being careful not to hurt
other people’s feelings (items 5, 7, 12, 24); Factor 9: Making requests to close
people (items 2, 8, 37, 40); Factor 10: Defending my consumer rights (items 1, 15,
35); Factor 11: Defending my own decisions before my parents (item 4). Items 3,
14, 17, 19 and 42 did not load above .40 on any factor, or did so on two or more
factors.
The questionnaire’s internal consistency (Cronbach’s alpha) was .88 and the
split-half reliability (Guttman) was .90.
 Assessing social skills with four self-report measures 387

Table 4
Exploratory factor analysis of the College Self-Expression Scale (CSES; Galassi et al., 1974)
and loadings of two representative items for each factor (n= 117)

Loadings of two CSES Factors

representative items for F1 F2 F3 F4 F5 F6 F7 F7 F9 F10 F11 ri-t
each factor of the CSES
Factor 1. Expressing annoyance, disgust, or displeasure (eigenvalue: 8.29; explained
variance: 16.57%)
48. If a friend unjustifiably
criticizes you, do you
.71 .22 -.05 -.02 .21 .20 .08 .12 -.02 .17 -.08 .59
express your resentment
there and then?
30. Do you express anger
or annoyance toward the
.58 .12 .06 .25 .10 .12 .21 -.09 .10 -.22 -.03 .43
opposite sex when it is
justified?
Factor 2. Expressing positive feelings (eigenvalue: 3.36; explained variance: 6.71%)
49. Do you express you
.04 .78 .22 .01 -.03 .15 .11 -.00 .06 .07 -.11 .39
feeling to others?
20. Are you able to express
love and affection to
.14 .71 .14 .11 .22 -.02 -.09 -.18 -.01 -.12 .04 .32
people for whom you
care?
Factor 3. Speaking in public or in front of the class (eigenvalue: 2.66; explained variance:
5.32%)
50. Do you avoid asking
questions in class for fear .17 .09 .80 .02 -.05 .12 .08 .10 .02 .12 .08 .54
of feeling self-conscious?
43. Do you freely volunteer
information or opinions in .06 .18 .78 -.01 .02 .02 .14 -.11 -.12 .11 .03 .41
class discussions?
Factor 4. Defending personal preferences before my parents (eigenvalue: 2.30; explained
variance: 4.59%)
29. If your parents want
you to come home for a
weekend buy you have
-.14 .04 .01 .66 .25 -.23 .16 -.03 -.05 .01 .09 .07
made important plans,
would you tell them of
your preference?
46. If your parents want
you home by a certain
time which you feel is
much too early and .27 .07 .11 .64 -.10 .19 .03 .11 .10 .02 .08 .39
unreasonable, do you
attempt to discuss or
negotiate this with them?
388 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

Factor 5. Giving compliments to friends (eigenvalue: 2.07; explained variance: 4.15%)

31. If a friend does and
errand for you, do you tell
.13 .11 .05 .06 .69 .01 .11 -.09 .01 -.11 -.11 .28
that person how much you
appreciate it?
25. If a friend is wearing a
new outfit that you like, .25 .30 -.08 .05 .60 .05 -.20 -.26 -.18 .10 .17 .27
do you tell that person so?
Factor 6. Interacting with the opposite sex (eigenvalue: 1.88; explained variance: 3.76%)
22. If a person of the
opposite sex whom you
have been wanting to
meet smiles of directs
.05 .14 .20 -.01 .07 .77 .05 .06 .05 -.02 -.10 .33
attention to you at a party,
would you take the
initiatives in beginning a
conversation?
44. Are you reluctant to
speak to an attractive
.22 -.01 .18 -.06 -.11 .56 -.06 -.19 .10 .07 .24 .27
acquaintance of the
opposite sex?
Factor 7. Defending my own rights before friends/roommates (eigenvalue: 1.73; explained
variance: 3.47%)
45. If you lived in an
apartment and the
landlord failed to make
.23 .02 .20 .05 .13 -.03 .61 .00 -.01 .02 -.03 .38
necessary repairs after
promising to do so, would
you insist on it?
16. If your roommate
without your permission
eats food that he or she
knows you have been .16 .12 .13 .21 -.09 .06 .52 -.10 .12 .21 -.06 .32
saving, can you express
your displeasure to you
roommate?
Factor 8. Being careful not to hurt other people’s feelings (eigenvalue: 1.70; explained
variance: 3.39%)
12. Are you overly careful
to avoid hurting other .04 -.24 .07 -.04 -.11 -.02 -.04 .62 .24 .25 .09 .13
people’s feelings?
7. Is it difficult for you to
compliment and praise .07 -.43 -.01 .08 -.20 -.09 .00 .54 -.25 -.12 -.27 .10
others?
Factor 9. Making requests to close people (eigenvalue: 1.59; explained variance: 3.17%)
37. Would you be hesitant
about asking a good friend .23 .02 -.11 .02 -.11 .28 .13 .19 .63 -.02 -.00 .25
to lend you a few dollars?
 Assessing social skills with four self-report measures 389

2. When you decide that

you no longer wish to date
someone, do you have
.10 .27 .24 .18 -.22 -.21 .05 -.00 .50 .19 .18 .33
marked difficulty telling
that person of you
decision?
Factor 10. Defending my consumer rights (eigenvalue: 1.49; explained variance: 2.98%)
15. If food that is not to
your satisfaction is served
in a restaurant, would you -.02 -.08 .26 .08 -.10 -.00 .05 -.04 .13 .72 -.11 .23
complain about it to the
waiter?
1. Do you ignore it when
someone pushes in front .28 .08 .13 -.11 .12 .02 .15 .22 -.03 .61 .07 .46
of you in line?
Factor 11. Defending my own decisions before my parents (eigenvalue: 1.36; explained
variance: 2.73%)
4. If you decide to change
your major to a field which
you parents will not -.01 .05 .09 .07 -.03 -.00 -.03 .02 .00 -.04 .82 .15
approve, would you have
difficulty telling them?
Note: CSES= College Self-Expression Scale; ri-t= item-total correlations.

Relationships across the various social skills questionnaires

Pearson’s correlations across the different questionnaires used in the present

study were calculated, including the new CHASO-I factors. Table 5 shows these
correlations with special emphasis on the high correlations between the total score
(sum of the 12 factors) of CHASO-I and the overall scores of the RAS (r= .66) and,
particularly, of the CSES (r= .77). The correlation between the CHASO-I and the
total score of the assertiveness or response probability subscale of the AI was
moderate (r= -.49). Regarding the relationships between the 12 CHASO-I factors
and the total score of the other three questionnaires, the highest correlations were
with the CSES, followed closely by the RAS. However, the correlations with the AI
ranged from moderate to low. In general, the AI was the self-report measure of
assertiveness with the weakest performance across the entire study.
The relationships among the three traditional questionnaires were also
calculated. The AI to RAS relationship was moderate (r= -.44) as it was between
the AI and the CSES (r= -.53). However, the relationship between the RAS and the
CSES was particularly strong (r= .77).
We also computed the relationships between the CHASO-I factors and those
factors named similarly across the other three social skills questionnaires (we
discarded CHASO-1 factors 5, 7 and 12 because they either appeared in broader
factors in the same questionnaire or lacked correspondence with any factor in the
other three questionnaires) (see table 6). Accordingly, we found that the factor
“Speaking or performing in public/Interacting with people in authority” was
included in the CHASO-I (factor 1) as well as in the CSES (factor 3), with a high
390 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

correlation of .77, as well as (at least partially) in the RAS (r= .54). We did not find
this factor in the AI.

Table 5
Correlations between the Social Skills Questionnaire-I (SSQ-I) and its factors, and the overall
score of the social skills questionnaires used in this study

SSQ-I total RAS AI-RP CSES

Factors of the SSQ-I and its total score score (12 total total total
factors) score score score
F1. Speaking or performing in public/ Interacting
.78 .62 -.42 .59
with people in authority
F2. Expressing positive feelings .72 .25 -.40 .44
F3. Refusing requests .56 .48 -.27** .52
F4. Interacting with persons I am attracted to .62 .40 -.22* .41
F5. Dancing or singing in public .41 .20 -.21* .35
F6. Disclosing information about myself to close
.58 .26 -.26** .30**
persons
F7. Asking an attendant or a stranger for
.64 .38 -.24* .39
something
F8. Expressing annoyance, disgust, or displeasure .65 .62 -.36 .55
F9. Expressing different opinions/Clarifying
.68 .44 -.35 .55
opinions
F10. Apologizing/Recognizing my own mistakes .59 .16** -.21* .34
F11. Interacting with strangers .79 .51 -.29** .53
F12. Thanking for congratulations/Arguing with
.46 .27 -.22* .33
salespeople
SSQ-I total score (12 factors) -- .66 -.49 .77
Note: SSQ-I= Social Skills Questionnaire-I; RAS= Rathus Assertiveness Schedule; AI-RP= Assertion
Inventory –Response probability; CSES= College Self-Expression Scale. All the correlations with p< .001,
except **p< .01, and *p< .05.

CHASO-I factor 2, “Expressing positive feelings” was also a factor in the other
three questionnaires cited here with a high correlation with the CSES (r= .65), and
a moderate one with the AI (r= -.56) and the RAS (r= .42). We should remember
that the negative correlation with the AI is due to the fact that the higher the
score in the AI, the lower the assertiveness or response probability.
The relationships between the rest of the CHASO-I factors and similar factors
in the other three measures are usually moderate, generally between .40 and .60,
although some CHASO-I factors do not pair with similar factors in all the other
questionnaires. On the other hand, the correlation among similar factors in the
three traditional social skills questionnaires also ranged from .39 to .59, except the
relationship between the RAS factor 2, “Expressing annoyance or displeasure”,
and factor 10 of the CSES, “Defending my consumer rights”, which had a high
correlation (r= .73).
 Assessing social skills with four self-report measures 391

Table 6
Relationships between the factors of the Social Skills Questionnaire-I and similar factors
from the other three social skills questionnaires

Similar factors in the traditional

SSQ-I Factors questionnaires
RAS AI-RP CSES
F1. Speaking or performing in public/Interacting F1= .54
-- F3= .77
with people in authority F4 = .57
F2. Expressing positive feelings F5= .42 F3= -.56 F2= .65
F3. Refusing requests F6= .41 -- F7= .52
F4. Interacting with persons I am attracted to F4= .48 -- F6= .49
F6. Disclosing information about myself to close
-- F3= -.42 --
persons
F8. Expressing annoyance, disgust, or F2= .57 F1= .47
F1= -.47
displeasure F6= .42 F10= .54
F9. Expressing different opinions/Clarifying F1= -.44 F1= .53
--
opinions F8= -.35 F3= .49
F10. Apologizing/Recognizing my own mistakes -- F7= -.43 --
F3= .59
F11. Interacting with strangers F4= .61 --
F6= .44
Note: SSQ-I= Social Skills Questionnaire-I; RAS= Rathus Assertiveness Schedule; RAS F1= Interacting
with commercial firms; RAS F2= Expressing annoyance or displeasure; RAS F4= Interacting with others
in an extroverted way; RAS F5= Expressing feelings openly; RAS F6= Refusing requests; AI-PR=
Assertion Inventory –Response Probability; AI-RP F1= Expressing annoyance, disgust or displeasure; AI-
RP F3= : Giving and receiving compliments, requests and asking personal questions; AI-RP F7=
Apologizing/Recognizing my own mistakes; AI-RP F8= Expressing different opinions/Clarifying opinions;
CSES= College Self-Expression Scale; CSES F1= Expressing annoyance, disgust, or displeasure; CSES F2=
Expressing positive feelings; CSES F3= Speaking in public or in front of the class; CSES F6= Interacting
with the opposite sex; CSES F7= Defending my own rights before friends/roommates; CSES F10=
Defending my consumer rights. Factors 5, 7, and 12 of the SSQ-I were not included in this table (see
text). All the correlations are statistically significant with p< .001.

Differences in social skills between men and women

We also calculated the differences between men and women in the factors
and in the total score of the CHASO-I and the other three social skills
questionnaires (RAS, AI, and CSES). There are few differences, and they are found
only in some of the CHASO-I factors, such as “Speaking or performing in
public/Interacting with people in authority”, “Interacting with persons I am
attracted to”, “Interacting with strangers” (with men scoring higher than women
in all of them) or “Apologizing/Recognizing my own mistakes” (with women
scoring higher than men). There were no sex differences in the total score for
three of the questionnaires. Only in the RAS total score were there statistically
significant differences (although with a low effect size, d= 0.43) between men and
women (with men scoring higher than women) (see table 7). Due to space limitations,
the sign of U has not been included.
392 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

Table 7
Differences between men and women in the four self-report measures of social skills

Factors and total scores for the various measures of Men Women
t p d
social skills M (SD) M (SD)
Social Skills Questionaire (SSQ-I)
F1. Speaking or performing in public/Interacting 37.70 34.90
2.77 .006 0.33
with people in authority (8.38) (8.64)
49.02 50.53
F2. Expressing positive feelings 1.59 .112 0.18
(8.40) (8.00)
21.00 20.55
F3. Refusing requests 0.88 .380 0.10
(4.28) (4.39)
21.67 17.89
F4. Interacting with persons I am attracted to 5.19 .000 0.58
(6.86) (6.05)
5.31 5.74
F5. Dancing or singing in public 1.95 .051 0.22
(1.92) (1.89)
F6. Disclosing information about myself to close 13.99 13.96
0.08 .938 0.01
persons (3.02) (3.03)
11.40 11.54
F7. Asking an attendant or a stranger for something 0.52 .606 0.06
(2.58) (2.32)
30.55 29.61
F8. Expressing annoyance, disgust, or displeasure 1.35 .177 0.16
(5.99) (5.98)
19.61 19.23
F9. Expressing different opinions/Clarifying opinions 1.03 .304 0.12
(3.46) (3.12)
31.80 33.13
F10. Apologizing/Recognizing my own mistakes 2.34 .020 0.25
(5.76) (4.58)
27.07 25.47
F11. Interacting with strangers 2.18 .030 0.26
(6.02) (6.34)
F12. Thanking for congratulations/Arguing with 6.50 6.52
0.11 .905 0.01
salespeople (1.74) (1.61)
273.94 270.00
SSQ-I total score (12 factors) 0.83 .405 0.10
(39.90) (37.20)
Rathus Assertiveness Schedule (RAS)
3.50 1.42
F1. Interacting with commercial firms 2.82 .005 0.33
(6.10) (6.44)
-0.35 -1.83
F2. Expressing annoyance or displeasure 2.34 .020 0.27
(5.37) (5.49)
0.89 -1.41
F4. Interacting with others in an extroverted way 3.02 .003 0.35
(6.81) (6.44)
-0.77 -2.04
F6. Refusing requests 2.33 .020 0.27
(4.85) (4.62)
10.08 1.43
RAS total score 3.35 .001 0.43
(18.39) (22.01)
Assertion Inventory (AI)
99.40 101.78
IA-RP total score 0.76 .449 0.11
(24.47) (17.69)
College Self-Expression Scale (CSES)
9.48 7.28
F3. Speaking in public or in front of the class 3.04 .003 0.66
(3.20) (3.42)
7.65 6.42
F6. Interacting with the opposite sex 2.13 .035 0.48
(2.27) (2.81)
131.00 127.63
CSES total score 0.74 .460 0.16
(19.61) (21.27)
 Assessing social skills with four self-report measures 393

Discussion

The assessment of social skills by means of self-report instruments has

remained almost unchanged for a good number of years. Questionnaires
developed in the 1970s are still being widely used and it appears that this situation
will hold true in the near future. In this study we have analyzed some of the
psychometric features of three of the self-report measures more widely used
around the world, namely, the Rathus Assertiveness Schedule (RAS), the Assertion
Inventory (AI), and the College Self-Expression Scale (CSES), and we have added a
questionnaire recently developed by our research team, the Social Skills
Questionnaire (SSQ-I). Considering the factorial structure issues related to this type
of measures, as stated here in the introduction, we decided to compare the
factorial structure of all four questionnaires and obtain their reliability and
convergent validity.
The results show that some social skills dimensions seem to be shared by the
four questionnaires, such as “expressing positive feelings”, “expressing
annoyance, disgust, or displeasure”, and “refusing requests”. Other dimensions
are shared by three of the questionnaires, such as “speaking or performing in
public/interacting with people in authority”, “interacting with persons I am
attracted to/of the opposite sex”, “interacting with strangers”, “expressing
different opinions/clarifying opinions”, or by only two, such as
“apologizing/recognizing my own mistakes” and “disclosing information about
myself to close persons”. Finally, some dimensions were present in only one of the
questionnaires, such as “dancing or singing in public” (SSQ-I), “defending my own
position” (RAS), “resisting undesirable propositions” (AI), “defending personal
preferences before my parents” (CSES), among others (see tables 1 to 4, and table
6).
Considering our own results as well as those in the literature regarding the
same topic (Furnham & Henderson, 1984; Galassi & Galassi, 1977; Gambrill &
Richey, 1975), there seem to be very few dimensions that are always present when
assessing social skills (see the dimensions shared by three or four questionnaires).
However, there would also seem to be a common discrepancy regarding the
number of dimensions or class of responses that the social skills construct should
contain; while we have found six dimensions in the RAS, there were eight in the
AI, and eleven in the CSES. Furthermore, such dimensions do not remain constant
for the same questionnaire, with their number and name changing in different
studies. Specifically, 11 and 13 factor solutions have been reported for the AI (e.g.,
Carrasco et al., 1989; Henderson & Furnham, 1983), whereas in this study we
found only eight. The RAS factorial solutions range from four to 12 factors (e.g.,
Heimberg & Harrison, 1980; Henderson & Furnham, 1983; Hull & Hull, 1978;
Nevid & Rathus, 1979), while in this study we have found six factors. Regarding
the CSES, the reported solutions have ranged from four to 14 factors (e.g., Galassi
& Galassi, 1973, 1979; Henderson & Furnham, 1983; Kipper & Jaffe, 1978), and in
this study we have found 11 factors, making it the only factorial solution matching
former research (Caballo & Buela, 1988).
394 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

Another frequent problem regarding social skills questionnaires is that many

of the items load on more than one factor, complicating the clarity and
interpretation of the factors. For instance, the study by Carrasco et al. (1989)
found that 25 of the 40 AI items loaded on more than one factor, and 15 of those
items even loaded on three or more factors. The new SSQ-I has kept only those
items loading on one factor and has maintained the dimensions with similar
corresponding factors in other traditional questionnaires (although many of the
items in the SSQ-I factors are formulated differently to those in the traditional
questionnaires).
Regarding the convergent validity of the SSQ-I, the correlation data for the
SSQ-I and the other questionnaires used to assess social skills were moderate,
supporting the notion that these self-report measures partially coincide with the
construct they are measuring, particularly in reference to the CSES and RAS. It
seems that the CSES has been considered the best questionnaire for assessing
social skills in university students (Burkhart, Green, & Harrison, 1979; Caballo,
1993; Jakubowski & Lacks, 1975), and the correlation with SSQ-I (r= .77) provides
major support for its validity. Regarding the dimensions of the SSQ-I, they also
seem to provide sound support for its validity through the correlations both with
the overall scores in traditional questionnaires and with the factors with similar
names in these selfsame questionnaires (see tables 5 and 6).
The AI and its assertiveness subscale have a weaker performance, with lower
correlations than the RAS or the CSES possibly because the AI contains items that
suit the adult population much better than university students, and our subjects
could be responding to future events, ones not yet faced, for instance “asking for
a raise”, “applying for a job”, or “quitting a job”. Furthermore, comparing the
SSQ-I to the AI suggest that the former more clearly separates the dimensions of
“expressing positive feelings” and “disclosing information about myself to close
persons” than the AI, which lumps them both together in a single factor (“giving
and receiving compliments, requests and asking personal questions”).
Regarding the CSES, it seems to be the best measure to evaluate university
students’ social skills, as mentioned earlier, but the SSQ-I may have two
advantages over the CSES. First, it does a better job of differentiating between two
dimensions such as “interacting with persons I am attracted to” and “interacting
with strangers”, while the CSES does not do so (maybe due to the way the CSES
combines the items). Second, the dimension “expressing different
opinions/clarifying opinions” seems to be better defined in the SSQ-I than in the
CSES, in which the items are mixed with other very different items in two different
factors.
On the other hand, some authors have considered that the RAS should
include items that measure aggressiveness rather than assertiveness (e.g., Galassi &
Galassi, 1975; Gritzmacher & Tucker, 1979; Heimberg & Harrison, 1980), such as
the item “There are times when I look for a good, vigorous argument”.
Furthermore, this questionnaire does not include items reflecting dimensions such
as “apologizing/recognizing my own mistakes”, “expressing different
opinions/clarifying opinions”, “disclosing information about myself to close
persons”, and even “expressing positive feelings” (considering this last dimension
 Assessing social skills with four self-report measures 395

with the comprehensive meaning of expressing affect, support, or giving

compliments) - all response classes that seem to belong to the social skills
construct. Comparing the SSQ-I with the RAS, we also find the former has two
advantages over the latter, whereby the SSQ-I better differentiates four
dimensions, on the one hand, “expressing annoyance, disgust, or displeasure” and
“refusing requests” and, on the other, “interacting with persons I am attracted
to” and “interacting with strangers”, which seems more complicated in the RAS,
with only one dimension, “Interacting with others in an extroverted way”, mixing
items from the other former four dimensions.
To sum up, regarding the convergent validity of the SSQ-I and the kind of
dimensions all the questionnaires included in this study measure, we may affirm
that the moderate relationships between the SSQ-I and all the other self-report
instruments support the validity of the new SSQ-I. However, our results also show
that the questionnaires do not assess the same dimensions of the social skills
construct, coinciding only partially in the assessment of some of its key features.
Regarding sex-related differences in the several social skills measures we
barely found any between men and women. The most relevant involve four
dimensions or response classes. The first of these dimensions is “Speaking or
performing in public/Interacting with people in authority”. Men score significantly
higher than women, with an effect size ranging low (SSQ-I, RAS) and moderate
(CSES). The sex related differences in this dimension are also found when
considering the social anxiety construct (Caballo et al., 2008, 2010; Caballo,
Salazar, Irurtia, Arias, et al., 2014), in which women scored higher than men in the
anxiety related to “Speaking in public/Talking with people in authority”,
differences that already seem to exist in puberty (Caballo, Arias, et al., 2012). The
overall data are consistent regarding this dimension, since a high social skill is
often associated with a low level of social anxiety (Caballo, Salazar, Irurtia,
Olivares, et al., 2014). In this regard, a possible explanation for the differences in
the skilled behavior of speaking in public or with authority figures may be related
to the cultural background and contexts surrounding such skills. Such behaviors
are nearly always part of formal education and work contexts - two types of
situations that demand these kinds of skills and traditionally involve mostly males.
It is only lately than females have had the chance to break with existing cultural
patterns of submission or inhibition.
The second dimension in which sex related differences were found is
“interacting with persons I am attracted to”. As in the first dimension, men have a
significantly higher score than women, and the effect size of these differences
varies from low (RAS, CSES) to moderate (SSQ-I). This dimension also behaves
similarly to the first one when we compare these data with those of a similar
dimension in the social anxiety construct (“Interaction with the opposite sex”). In
this case, females scored higher than males regarding the anxiety related to the
“Interaction with the opposite sex” (see Caballo et al., 2008, 2010; Caballo,
Salazar, Irurtia, Arias, et al., 2014), differences that also seem to exist in puberty
(Caballo, Arias, et al., 2012). In order to explain why males seem to be more
assertive when relating to people they are attracted to, we again refer to the
possible influence of sociocultural norms that are still transmitted from one
396 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

generation to another in our society. We should remember that, culturally

speaking, we have assigned a more active role to males and a more passive one to
females when it comes to searching for a mate. While it is true that certain subtle
pattern changes can be observed today regarding flirting and courtship, our data
suggest that there are still sex differences giving males (more often than females)
the initiative in this kind of behaviors.
The third dimension in which there were sex differences was “Interacting
with strangers.” Again, men scored significantly higher than women with a low
effect size of the differences (SSQ-I, RAS). As in the two previous dimensions, there
is a counterpart dimension in the social anxiety construct called “Interaction with
strangers” in which women scored higher than men (Caballo et al., 2008, 2010;
Caballo, Salazar, Irurtia, Arias, et al., 2014), although these differences are also
much lower than in the first two dimensions.
The fourth dimension of social skills in which sex differences were found was
“Apologizing/Recognizing my own mistakes”. In contrast with the first three
dimensions, women scored significantly higher than men in this fourth one
(although with a low effect size), and were therefore more skillful in this
dimension. At the risk of sounding like a scratched record, we will again resort to
socialization and cultural norms as a possible explanation for these differences,
since apologizing and recognizing our own mistakes have often been assumed as
weaknesses in Western society, which is incompatible with the (more aggressive)
“male image” but not with the (more submissive) “female image”.
In regard to the remaining dimensions the data are not so clear, so while
there are differences (of low effect size) between males and females in some of
the dimensions assessed by a given questionnaire (e.g., “Expressing annoyance or
displeasure” in the RAS), this does not happen in terms of this same dimension in
other questionnaires. When considering the total scores in the questionnaires, the
only significant difference found between males and females was in the RAS score
(which may also be assessing aggressiveness, as we said before), but no significant
differences were found in the other three questionnaires. In conclusion, it seems
that here might be significant differences associated with sex in the four kinds of
social skills listed above, but when we consider only the overall score in the
questionnaires such differences may be masked and will not appear in the results.
Finally, we would like to note certain limitations of this study. First, the
participant sample focused mainly on university students, so the results may not be
generalizable to other populations. Secondly, one of the questionnaires is a pilot
version, not the final one, so the results might end up differing. However, in spite
of these limitations, we believe that this study may help to further clarify the issue
of how many and what dimensions or response classes the social skills construct
might include. Furthermore, it is also an important step in the development of a
new and comprehensive self-report measure of social skills, as is the “Social Skills
Questionnaire” (SSQ-I), although there is still a long way to go.
 Assessing social skills with four self-report measures 397

References

Beck, J. G., & Heimberg, R. G. (1983). Self-report assessment of assertive behavior: A critical
analysis. Behavior Modification, 7, 451 487.
Burkhart, B. R., Green, S. B., & Harrison, W. H. (1979). Measurement of assertive behavior:
Construct and predictive validity of self-report, role-playing and in-vivo measures.
Journal of Clinical Psychology, 35, 376-383.
Caballo. V. E. (1986). Evaluación de las habilidades sociales. In R. Fernández Ballesteros, & J.
A. Carrobles (Eds.), Evaluación conductual: Métodos y aplicaciones (3th ed.) (pp. 553-
595). Madrid: Pirámide.
Caballo, V. E. (1993). Relationships among some behavioral and self-report measures of
social skills. Psicología Conductual, 1, 69-91.
Caballo, V. E. (1997). Manual de evaluación y entrenamiento de las habilidades sociales (2nd
ed.). Madrid: Siglo XXI.
Caballo, V. E., & Buela, G. (1988). Factor analyzing the College Self-Expression Scale with a
Spanish population. Psychological Reports, 63, 503-507.
Caballo, V. E., Salazar, I. C., Arias, B., Irurtia, M. J., Calderero, M., & CISO-A Research Team
Spain (2010). Validation of the Social Anxiety Questionnaire for Adults (SAQ-A30) with
Spanish university students: Similarities and differences among degree subjects and
regions. Behavioral Psychology/Psicología Conductual, 18, 5-34.
Caballo, V. E., Arias, B., Salazar, I. C., Calderero, M., Irurtia, M. J., & Ollendick, T. H. (2012).
A new self-report assessment measure of social phobia/anxiety in children: The Social
Anxiety Questionnaire for Children (SAQ-C24). Behavioral Psychology/Psicología
Conductual, 20, 485-503.
Caballo, V. E., Salazar, I. C., & Irurtia, M. J. (2014). Social Skills Questionnaire (SSQ-I)
(“Cuestionario de habilidades sociales”, CHASO-I). Unpublished manuscript.
Caballo, V. E., Salazar, I. C., Irurtia, M. J., Arias, B., Hofmann, S. G., & CISO-A Research
Team (2008). Social anxiety in 18 nations: Sex and age differences. Behavioral
Psychology/Psicología Conductual, 16, 163-187.
Caballo, V. E., Salazar, I. C., Irurtia, M. J., Arias, B., Hofmann, S. G., & CISO-A Research
Team (2012). The multidimensional nature and multicultural validity of a new measure
of social anxiety: The Social Anxiety Questionnaire for Adults. Behavior Therapy, 43,
313-328.
Caballo, V. E., Salazar, I. C., Irurtia, M. J., Arias, B., Hofmann, S. G., & CISO-A Research
Team (2014). Differences in social anxiety between men and women across 18
countries. Personality and Individual Differences, 64, 35-40.
Caballo, V. E., Salazar, I. C., Irurtia, M. J., Arias, B., & Nobre, L. (2013). The assessment of
social anxiety through five self-report measures, LSAS-SR, SPAI, SPIN, SPS, and SIAS: A
critical analysis of their factor structure. Behavioral Psychology/Psicología Conductual,
21, 423-448.
Caballo, V. E., Salazar, I. C., Irurtia, M. J., Olivares, P., & Olivares, J. (2014). The relationship
between social skills and social anxiety and personality styles/disorders. Behavioral
Psychology/ Psicología Conductual, 22, 401-422.
Carmona, A. E., & Lorr, M. (1992). Dimensions of assertiveness: A cross-cultural comparison
of Chilean and U.S. subjects. Personality and Individual Differences, 13, 45-48.
Carrasco, I., Clemente, M., & LLavona, L. (1983). Análisis de componentes básicos del
Inventario de asertividad de Rathus. Revista Española de Terapia del Comportamiento,
1, 249-264.
Carrasco, I., Clemente, M., & Llavona, L. (1989). Análisis del Inventario de aserción de
Gambrill y Richey. Estudios de Psicología, 37, 63-74.
398 CABALLO, SALAZAR, OLIVARES, IRURTIA, OLIVARES, AND TOLEDO

Casas-Anguera, E., Prat, G., Vilamala, S., Escandell, M. J., García-Franco, M., Martín, J. R.,
López, E., & Ochoa, S. (2014). Validación de la versión española del Inventario de
asertividad Gambrill y Richey en población con diagnóstico de esquizofrenia. Anales de
Psicología, 30, 431-437.
Castaños Cervantes, S., Reyes Lagunes, I., Rivera Aragón, S., & Díaz loving, R. (2011).
Estandarización del Inventario de asertividad de Gambrill y Richey-II. Revista
Iberoamericana de Diagnóstico y Evaluación Psicológica, 29, 27-50.
Curran, J. P. (1979). Pandora's box reopened? The assessment of social skills. Journal of
Behavioral Assessment, 1, 55-71.
Del Prette, Z. A. P., Del Prette, A., & Barreto, M. C. M. (1998). Análise de um Inventário de
Habilidades Sociais (IHS) em uma amostra de universitários. Psicologia: Teoria e
Pesquisa, 14, 219-228.
Furnham, A., & Henderson, M. (1984). Assessing assertiveness: A content and correlational
analysis of five assertiveness inventories. Behavioral Assessment, 6, 79-88.
Galassi, J. P., DeLo, J. S., Galassi, M. D., & Bastien, S. (1974). The College Self-Expression
Scale: A measure of assertiveness. Behavior Therapy, 5, 164-171.
Galassi, J. P., & Galassi, M. D. (1973). A factor analysis of a measure of assertiveness.
Unpublished manuscript, West Virginia University, USA.
Galassi, J. P., & Galassi, M. D. (1975). Relationship between assertiveness and
aggressiveness. Psychological Reports, 36, 352-354.
Galassi, J. P., & Galassi, M. D. (1977). Assessment procedures for assertive behavior. En R. E.
Alberti (Ed.), Assertiveness: Innovations, applications, issues (pp. 307-325). San Luis
Obispo, CA: Impact.
Galassi, J. P., & Galassi, M. D. (1979). A comparison of the factor structure of an assertion
scale across sex and population. Behavior Therapy, 10, 117-128.
Gambrill, E. D., & Richey, C. A. (1975). An assertion inventory for use in assessment and
research. Behavior Therapy, 6, 550-561.
Gay, M. L., Hollandsworth, J. G. Jr., & Galassi, J. P. (1975). An assertiveness inventory for
adults. Journal of Counseling Psychology, 22, 340-344.
Gritzmacher, K. K., & Tucker, R. K. (1979, May). A multivariate investigation of selected
assertiveness instruments. Paper presented at the annual convention of the
International Communication Association, Philadelphia, USA.
Heaven, P. C. L. (1984). Factor structure of the Lorr and More Assertiveness Inventory.
Personality and Individual Differences, 5, 741-742.
Heimberg, R. G., & Harrison, D. F. (1980). Use of the Rathus Assertiveness Schedule with
offenders: A question of questions. Behavior Therapy, 11, 278-281.
Henderson, M., & Furnham, A. (1983). Dimensions of assertiveness: Factor analysis of five
assertion inventories. Journal of Behavior Therapy and Experimental Psychiatry, 14,
223-231.
Hersen, M., Bellack, A. S., Turner, S. M., Williams, M. T., Harper, K., & Watts, J. G. (1979).
Psychometric properties of the Wolpe-Lazarus Assertiveness Scale. Behaviour Research
and Therapy, 17, 63-49.
Hull, D. B., & Hull, J. H. (1978). Rathus Assertiveness Schedule: Normative and factor-
analytic data. Behavior Therapy, 9, 673.
Jakubowski, P. A., & Lacks, P. B. (1975). Assessment procedures in assertion training. The
Counseling Psychologist, 5, 84-90.
Kipper, D. A., & Jaffe, Y. (1978). Dimensions of assertiveness: Factors underlying the College
Self-Expression Scale. Perceptual and Motor Skills, 46, 47-52.
Kogan, E. S., Hersen, M., Kabacoff, R. I., & Van Hasselt, V. B. (1996). Psychometric
properties of the Wolpe-Lazarus Assertiveness Scale with community-dwelling older
adults. Journal of Psychopathology and Behavioral Assessment, 17, 97-109.
 Assessing social skills with four self-report measures 399

LaFromboise, T. D. (1983). The factorial validity of the Adult Self­ Expression Scale with
American Indians. Educational and Psychological Measurement, 43, 547-555.
Lazarus, A. A. (1973). On assertive behavior: A brief note. Behavior Therapy, 4, 697-699.
Lorr, M., & More, W. W. (1980). Four dimensions of assertiveness. Multivariate Behavioral
Research, 15, 127-138.
Nevid, J. S., & Rathus, S. A. (1979). Factor analysis of the Rathus Assertiveness Schedule
with a college population. Journal of Behavior Therapy and Experimental Psychiatry,
10, 21-24.
Olaz, F. O., Medrano, L., Greco, M. E., & Del Prette, Z. A. P. (2009). Argentinean adaptation
of the Social Skills Inventory IHS-Del Prette. The Spanish Journal of Psychology, 12,
756-766.
Olivares, J., Olivares-Olivares, P. J., & Macià, D. (2014). Entrenamiento en habilidades
sociales y tratamiento de adolescentes con fobia social generalizada. Behavioral
Psychology/Psicología Conductual, 22, 441-459.
Rathus, S. A. (1973). A 30-item schedule for assessing assertive behavior. Behavior Therapy,
4, 398-406.
Statsoft (2013). Statistica, v. 12 (statistical program). Tulsa, OK: Statsoft.
Stevens, J. (1986). Applied multivariate statistics for the social sciences. Hillsdale, NJ:
Erlbaum.
Vaal, J. (1975). The Rathus Assertiveness Schedule: Reliability at the junior high school level.
Behavior Therapy 6, 566-567.
Wolpe, J., & Lazarus, A. A. (1966). Behavior therapy techniques: A guide to the treatment
of neurosis. Oxford: Pergamon.

RECEIVED: May 5, 2014

ACCEPTED: July 10, 2014

View publication stats