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

net/publication/232575173

Assessment of Basic Social Skills

Article in Journal of Personality and Social Psychology · September 1986

DOI: 10.1037/0022-3514.51.3.649

CITATIONS READS
927 34,288

1 author:

Ronald E. Riggio
Claremont McKenna College
812 PUBLICATIONS 14,569 CITATIONS

SEE PROFILE

All content following this page was uploaded by Ronald E. Riggio on 23 April 2015.

The user has requested enhancement of the downloaded file.

Journal of Personality and Social Psychology Copyright 1986 by the American Psychological Association, Inc.
1986, Vol.51, No. 3,649-660 0022-3514/86/$OC.7S

Assessment of Basic Social Skills

Ronald E. Riggio
California State University, Fullerton

Following recent developments in the measurement of individual differences in nonverbal social

skills, we proposed a conceptual framework for defining and assessing basic social skills. Preliminary
testing resulted in the development of a 105-item, pencil-and-paper measure of seven basic dimen-
sions of social skills, called the Social Skills Inventory (SSI). In a series of validation studies using
undergraduate students, the SSI demonstrated convergent and discriminant validity in relation to
other measures of nonverbal social skill and traditional personality scales. Scores on the SSI also
predicted some social group memberships, typical social behaviors, and the depth of social networks.
This evidence suggests that the SSI could prove to be a valuable tool for research in personality and
social psychology and for work in applied settings.

In recent years there has been increased attention to denning recent are the attempts to measure individual differences in
and assessing individual differences in social abilities and inter- nonverbal communication skill (Rosenthal, 1979; see also
personal skills. Work in this area is progressing on several O'Sullivan, 1983, for a historical overview).
fronts. Psychologists have become increasingly concerned with Although the present research is most closely aligned with
the assessment and development of social skills for assisting this latest orientation, it is impossible to define social skills
clinical populations (see Curran & Monti, 1982; Trower, Bry- without incorporating the work of a vast array of social science
ant, & Argyle, 1978; Wine & Smye, 1981). Personality and so- researchers. There is perhaps no adequate single definition of
cial psychologists have developed standardized instruments that social skills. The variety and assortment of dimensions labeled
assess dimensions related to interpersonal skill and social effec- as social skills is enormous. Yet there are certain consistencies
tiveness, for example, measures of constructs such as empathy in the varying theoretical perspectives. Many social skill re-
(Hogan, 1969; Mehrabian & Epstein, 1972), shyness, sociabil- searchers agree that the basic sending and receiving of informa-
ity, (Cheek & Buss, 1981), and self-monitoring (Snyder, 1974, tion represent key social skills. Indeed, Hall (1979) divided so-
1979). Communication researchers have concerned themselves cial communication skills into two broad classes of sending and
with the assessment of communicative competence (Diez, receiving. Additional social skills involve cognitive abilities
1984;Wiemann, 1977;Wiemann&Backlund, 1980). Although such as interpersonal problem-solving skills and role-playing
these represent different research perspectives, they may all be abilities (Meichenbaum, Butler, & Gruson, 1981).
attempting to tap into the same core dimensions of social skill. Many existing measures of social skills focus on a single, spe-
Attempts to measure individual differences in social abilities, cific type of skill, skill deficit, or skill-related construct such as
or skills, have their roots in the work of Thorndike and others assertiveness (Rathus, 1973), nonverbal sensitivity (Rosenthal,
on social intelligence (e.g., Hunt, 1928; E. L. Thorndike, 1920; Hall, DiMatteo, Rogers, & Archer, 1979), fear of negative evalu-
R. L. Thorndike, 1936; R. L. Thorndike & Stein, 1937). How- ation (Watson & Friend, 1969), or communication apprehen-
ever, difficulties in assessing social intelligence, particularly the sion (McCroskey, 1977). Some instruments that purport to
inability to discriminate social intelligence from general intelli- measure singular dimensions of social skill may, in fact, be as-
gence, led to the demise of this line of research. It was many sessing constructs that are truly multidimensional; that is, com-
years later that research on the measurement of social abilities prised of more basic independent social skills. Such appears to
was revived with the work of Ouilford and his colleagues on be the case with Snyder's Self-Monitoring Scale (SMS, Snyder,
behavioral intelligence (Guilford, 1967) and the development of 1974), which may be composed of three more basic social skills
scales to assess empathy (Dymond, 1949; Hogan, 1969). More (Briggs, Cheek, & Buss, 1980; Riggio & Friedman, 1982; see
also Lennox & Wolfe, 1984). Similar multidimensionality may
underly the constructs of empathy (Davis, 1983) and assertive-
This research was supported by intramural grants from California ness (Galassi, Galassi, & Vedder, 1981).
State University, Fullerton (CSUF) and from a CSUF President's Sum- The present study is an attempt to develop a general frame-
mer Research Grant. work for several basic dimensions of social skill and to report
Special thanks go to Barbara Throckmorton, Kathy Lang, and Bruce on the construction of a self-report assessment tool to measure
Smith for their tremendous assistance in data collection and to Maria these basic skill dimensions. This framework is derived from
Hale, Patti Hopkinson, Larisa Lamb, Mary Lybeck, Kevin McNulty,
multidisciplinary research on social and interpersonal skills,
Mitch Okada, and Debbie White for their help. Chris Cozby, Richard
but it springs most directly from the attempts of social personal-
Lippa, Keith Widaman, Stan Woll, and Judy Zimmerman made many
ity psychologists to measure individual differences in nonverbal
helpful comments and suggestions.
communication skills. Most notable of these attempts are the
Correspondence concerning this article should be addressed to Ron-
ald E. Riggio, Department of Psychology, California State University, work of Rosenthal and his colleagues (Rosenthal et al., 1979)
Fullerton, California 92634. and Buck (1984) on measuring nonverbal sensitivity, Friedman

649
650 RONALD E. RIGGIO

and his colleagues' work with nonverbal expressiveness (Fried- ing ability. In other cases relations among certain skill dimen-
man, Prince, Riggio, & DiMatteo, 1980; Friedman & Riggio, sions may tend to be negative. Such might be the case for the
1981), and Snyder's research on self-monitoring (Snyder, 1974, individual exceptionally skilled in emotional control who has
1979; see also Lennox & Wolfe, 1984). Each of these lines of some difficulty in the spontaneous expression of felt affective
research deals primarily with a separate dimension of commu- states. There is no doubt that the interrelations among the basic
nication skill—sensitivity to communication, communication components of social skill are complex.
expressivity, and monitoring or control of communication. The present definition of basic social skills assumes that such
These are the three primary skills upon which the present skills are learned social abilities and strategies. Thus, the term
framework is built. skill is used broadly. Exactly what constitutes a particular social
Skills in sending and receiving information are represented ability or skill is governed by the particular social and cultural
in the basic social skills of expressivity and sensitivity. The re- norms that regulate interpersonal communication. Generally
search cited above has dealt primarily with the communication to possess greater social skill is socially advantageous and may
of nonverbal messages, especially the sending and receiving of be, to a certain extent, socially desirable. Yet, social skill is not
emotions. Whereas emotional sending and receiving ability rep- a single entity but, rather, a constellation of many more basic
resent important components of nonverbal social skills, verbal skills. The relation between any single social skill dimension
encoding and decoding abilities are two separate, yet related, and social effectiveness (i.e., positive or desirable social out-
social skills. Thus, the first four social skills in our framework comes) may not always be linear. Possessing too much of any
are termed: emotional expressivity, emotional sensitivity, social one of the basic components of social skill, in relation to other
expressivity, and social sensitivity. It is important to note that key social skill components, may be dysfunctional. For exam-
although two of these dimensions are labeled emotional expres- ple, individuals high in expressivity but lacking skills in regula-
sivity and emotional sensitivity, they include more than just the tion and control may initially attract positive attention and re-
ability to communicate emotions. Emotional expressivity in- sponses from others, but they may soon be viewed as overly
volves skill in communicating affect, attitudes, and status. Emo- long-winded, frivolous, and ill-mannered. Persons exceedingly
tional sensitivity is the ability to decode others' emotions, be- high in social control but lacking in other important skills may
liefs, or attitudes, and cues of status-dominance. On the other become "social chameleons," able to adapt to just about any
hand, social expressivity includes skill in verbal expression, ver- social situation but unable to express their own feelings and
bal fluency, and ability to initiate conversations. Ability to re- perhaps unable to establish emotional ties with others. There-
ceive and understand verbal messages and knowledge and con- fore, although we can speak of specific components of general
cern for social rules and norms are central components of social social skill, these components become most important when
sensitivity. combined with other skill components. Amount or degree of
Two additional skills are concerned with control over com- each social skill dimension is important but so, too, is possess-
munication. The skill of emotional control is the ability to regu- ing a balance of the various social abilities.
late emotional communications and nonverbal displays. Social
control includes role-playing ability, regulation of verbal behav-
ior, and self-presentational skill. Measuring
In summary, the six basic social skill dimensions considered
in the present framework involve skill in sending, receiving, and Several standardized measures of basic social skills exist. Of
controlling communication in two separate domains, the emo- particular interest are instruments designed to assess certain
tional-nonverbal and the social-verbal domains. More detailed nonverbal skills. Some of these instruments are performance
outlines of these six basic social skill dimensions will be pre- measures of nonverbal social skills. The Profile of Nonverbal
sented later. Sensitivity (PONS; Rosenthal et al., 1979), the Communication
It is important to mention that these basic skill dimensions of Affect Receiving Ability Test (CARAT; Buck, 1976,1983), and
are meant to represent a general breakdown of social-commu- the Social Interpretations Test (SIT; Archer & Akert, 1977) are
nication skills. It is in no way an exhaustive list. Moreover, the performance measures of skill in decoding nonverbal commu-
basic skill dimensions outlined above may themselves be further nication. A number of techniques have also been developed to
broken down into specific subskills or abilities. However, this assess ability to spontaneously express emotions (e.g., Buck,
article will deal primarily with the general social skill frame- 1978; Zuckerman, Hall, DeFrank, & Rosenthal, 1976) and pose
work and validation of scales to assess the basic skill dhnen- emotional expressions (e.g., Friedman et al., 1980; Zuckerman,
Larrance, Hall, DeFrank, & Rosenthal, 1979). Friedman
(1979) asserted that such performance measures of nonverbal
social skills are reliable and valid methods of assessing key per-
Basic Social Skills
sonality dimensions. However, such performance measures are
costly and time consuming to administer. Attempts have been
Interrelationships
made to develop paper-and-pencil, self-report measures of some
Although in the present framework social skill is broken of these same nonverbal skills. The Affective Communication
down into six basic components, in most cases it seems likely Test (ACT; Friedman et al., 1980) is a standardized measure of
that development of one area of social skill may be positively spontaneous nonverbal expressiveness. Zuckerman and Lar-
related to development of certain other social skills. For exam- rance (1979) constructed self-report instruments to assess per-
ple, improvement in ability to decode the emotional expres- ceived encoding and decoding abilities; however, the validity of
sions of others may lead to the development of emotional send- these scales is questionable (see Riggio, Widaman, & Friedman,
 BASIC SOCIAL SKILLS 651

1985). Finally, Snyder's (1974) SMS, in part, measures skill in strong, felt emotions, thus controlling against the display of
nonverbal control. spontaneous and extreme emotional states.
Each of these instruments attempts to measure a single di- Social Expressivity (SE). Social expressivity refers to a gen-
mension of communication skill. The goal of the present study eral verbal speaking skill and an ability to engage others in social
was to develop a much needed general scheme for assessing cer- interaction. Persons high in SE appear outgoing and gregarious
tain key dimensions that combine to make up a large part of because of their ability to initiate conversations with others. So-
global social skill. This article reports on the development of a cially expressive individuals are usually able to speak spontane-
self-report inventory designed to assess seven basic social skill ously, sometimes without apparent control or monitoring of
dimensions. The inventory was intended to provide a relatively content.
short, easy-to-administer instrument for assessing individual Social Sensitivity (SS). The ability to decode and understand
differences in particular social abilities or skills and to serve as verbal communication and general knowledge of the norms
a tool for further research in the area. governing appropriate social behavior is termed SS. Socially
It is important to note here that self-report instruments of sensitive individuals are attentive to others (i.e., good watchers
communication skill (or skill-related constructs) do not usually and listeners). Because of their knowledge of social norms and
directly assess communication skills. Rather, these instruments rules, persons high in SS may become overconcerned with the
indirectly assess the basic communication abilities that com- appropriateness of their own behavior and the behavior of oth-
prise social skill. ers. In extremes, high socially sensitive persons' concern over
appropriate social behavior may lead to self-consciousness and
social anxiety, which may inhibit the persons' participation in
Dimensions social interaction.
Social Control (SC). Social control refers to a general skill
In order to provide a clearer picture of the theory underlying in social self-presentation. Individuals high in SC are tactful,
each of the proposed basic social skill dimensions, we will socially adept, and self-confident. Persons high in SC are skilled
briefly discuss each skill. at acting—able to play various social roles and can easily take
Emotional Expressivity (EE). Emotional expressivity refers a particular stance or orientation in a discussion. Individuals
to general skill in nonverbal sending. Some previous attempts high in SC are socially sophisticated and wise. Consequently
at measuring individual differences in emotional sending ability they are able to adjust personal behavior to fit with what they
have focused on posed emotional sending (e.g., Zuckerman, Li- consider to be appropriate to any given social situation.
pets, Koivumaki, & Rosenthal, 1975). However, this dimension A final dimension that was explored in the present study is
reflects individuals' ability to express, spontaneously and accu- termed Social Manipulation (SM). Rather than simply being a
rately, felt emotional states as well as the ability to nonverbally social ability, SM is also a general attitude or orientation. Per-
express attitudes and cues of interpersonal orientation. This sons high in SM believe that in certain social situations it is
definition of EE is more in line with that of Friedman et al. necessary (and useful) to manipulate others or alter elements of
(1980). Persons skilled in EE are animated and energetic and the situation to affect the outcome of social encounters. The SM
may be characterized as emotionally charged. Individuals who dimension was derived, in part, from Christie and Geis's (1970)
are high in EE may be able to emotionally arouse or inspire work on Machiavellianism and from research examining decep-
others because of their ability to transmit their felt emotional tion ability (i.e., DePaulo and Rosenthal, 1979; Riggio & Fried-
states (cf. Friedman & Riggio, 1981). Individuals in emotional man, 1983). Machiavellians are impersonal manipulators who
expressivity may tend to lack emotional control, owing to their use their social skills to further their own ends. However, unlike
emotional spontaneity. Machiavellians, high social manipulators do not simply try to
Emotional Sensitivity (ES). Emotional sensitivity refers to achieve social outcomes favorable to themselves. Persons high
general skill in receiving and decoding the nonverbal communi- in SM are simply willing, and able, to affect the outcomes of
cations of others. As such, it is closely related to nonverbal sen- social interaction. In certain instances, individuals high in SM
sitivity as denned by Rosenthal et al. (1979). Individuals high in may even act in a self-sacrificing manner (e.g., taking blame or
ES are concerned with and vigilant in observing the nonverbal responsibility in order to protect another). Thus, the concept of
emotional cues of others. Because persons high in ES are able to SM is distinct from Machiavellianism and may have important
decode emotional communication rapidly and efficiently, they implications for the use of social skills in influencing others.
may be more susceptible to becoming emotionally aroused by
others—sympathetically experiencing the emotional states of
Development of the Social Skills Inventory
others (see Friedman & Riggio, 1981).
Emotional Control (EC). The general ability to control and Item Generation and Pilot Testing
regulate emotional and nonverbal displays is referred to as EC.
The individual high in EC is likely to be a good emotional actor, On the basis of broad and detailed definitions of the seven
able to pose emotions on cue, and able to use conflicting emo- basic social skills, approximately 50 self-report items were gen-
tional cues to mask felt emotional states (i.e., laughing appro- erated for each of the seven dimensions. Although most of the
priately at a joke; putting on a cheerful face to cover sadness). items were written to assess a particular basic social skill dimen-
Emotional control may be one of the critical skills that, com- sion, some of the items are variations of items drawn from vari-
bined with other skills, is what Snyder (1974) referred to as self- ous self-report personality scales that seemed to assess some
monitoring (see Riggio & Friedman, 1982). The high emotion- component of social skill. Attempts were made to construct or
ally controlled person may tend to moderate the display of use items that would maximize the respondent's ability to accu-
652 RONALD E. RIGGIO

rately report aspects of their social skills (see Riggio, Widaman, Table 1
& Friedman, 1985). That is, many items related not to how the Sample Items From Each of the Social Skills
respondents perceived themselves but to how others perceived Inventory (SSI) Scales"
them. For example, items such as "People have told me that I
am a sensitive and understanding person" were used rather than Emotional Expressivity
When I get depressed, I tend to bring down those around me.
"I am a sensitive and understanding person." Whenever possi- I have been told that 1 have "expressive" eyes.
ble, we tried to use items that involved skills about which the Quite often I tend to be the "life of the party."
respondent had received some performance feedback from the
Emotional Sensitivity
social environment. It is nearly impossible for people to hide their true feelings from me.
Most items were written in both a forward and reversed for- At parties I can instantly tell when someone is interested in me.
mat. Pilot testing was done on both formats. The resulting pre- People often tell me that 1 am a sensitive and understanding person.
liminary version of the Social Skills Inventory (SSI) contained Emotional Control
138 items and had 9-point response scales with the anchors not I am able to conceal my true feelings from just about anyone.
at all true of me and very true of me. I am very good at maintaining a calm exterior, even when upset.
Two pilot samples of 48 and 42 undergraduate students were When I am really not enjoying myself at some social function, I can
still make myself look as if I am having a good time.
administered the preliminary version of the SSI in large groups.
Items were eliminated from this early version of the SSI if the Social Expressivity
distribution of responses was too negatively or positively skewed At parties I enjoy speaking to a great number of different people.
When in discussions, I find myself doing a large share of the talking.
(M < 2.8 or M > 7.2) or if the items showed too little variance
I usually take the initiative and introduce myself to strangers.
(SD < 1.9). A second version of the SSI (125 items) was con-
structed. It contained fewer and more refined items and was Social Sensitivity
Sometimes I think that I take things that other people say to me too
used in one of the validity studies. A final version of the SSI
personally.
contained some minor changes (i.e., deletion of some items) and I often worry that people will misinterpret something that I have said
improvements (i.e., rewording of items) from the second ver- to them.
sion and was used in the majority of the validity studies re- While growing up, my parents were always stressing the importance of
good manners.
ported here.
Social Control
The Social Skills Inventory I find it very easy to play different roles at different times.
When in a group of friends, I am often spokesperson for the group.
The SSI is a self-report, pencil-and-paper measure of seven I can fit in with all types of people, young and old, rich and poor.
basic social abilities. The entire inventory contains 105 items, Social Manipulation
15 items in each of the seven social skill subscales. The response I am not always able to tell the truth.
scale for each item is a 9-point scale containing corresponding If I really have to, I can "use" other people to get what I want.
numbers that range from —4 to +4 and with which the respon- Sometimes I feel that the social rules that govern other people don't
really apply to me.
dents indicate the extent to which each statement is true or false
as it applies to them (scale anchors: not at all true of me and • The complete SSI and additional reliability and validity information
very true of me). can be obtained by writing to the author.
The SSI consists of seven basic social skill scores, formed by
summing the 15 items for each subscale. Thirty-seven of the
105 items are worded in the reverse direction. Sample items
students, ranged from .75 to .88 and are shown in Table 2. These
from each subscale are presented in Table 1.
coefficients demonstrated the high internal consistency of all
The norms for the various SSI subscales and the scores for a
the SSI scales, and they compared favorably with other social
large sample are presented in Table 2.
skill instruments (e.g., Friedman etal., 1980;Snyder, 1974).
In addition to the seven basic skill scores, an index of global
Test-retest reliability. Forty undergraduate students were ad-
social skill can be computed by summing the seven scales. As
ministered the SSI on two occasions, 2 weeks apart. The test-
mentioned earlier, it is difficult to speak of a single index of
retest correlations ranged from .81 to .96 (all ps < .001) and are
social skill because it is a balance of social skill components
presented in Table 2.
that is critical in determining a truly socially skilled individual.
In sum, the SSI is composed of seven scales, each demonstrat-
Despite this, we used a total SSI score in many of these validity
ing high reliability, comparable to similar, short, self-report
studies. Ideally, we should have eliminated from our ranks of
scales.
socially skilled individuals those with extreme imbalances in
Intercorrelations among SSI scales. As mentioned earlier,
social skill component scores. However, we decided to take the
having a particular social skill is expected in most cases to be
conservative route and treat total SSI as a continuous variable,
positively intercorrelated with possessing other social skills, be-
feeling that total score of the SSI was a robust variable. This
cause one social skill usually predisposes an individual to de-
was a cautious strategy in that we may have underestimated any
velop other skills. For example, the ability to accurately recog-
relation that existed.
nize emotional expressions in others (ES) may eventually lead
to a better ability to express emotions. Thus, we usually speak
Reliability of certain individuals who are socially skilled (or not socially
Internal consistencies. Coefficient alphas for the seven SSI skilled) on a variety of dimensions. However, the possession of
scales, which were derived from a sample of 149 undergraduate one skill may preclude, in certain cases, having another social
 BASIC SOCIAL SKILLS 653

Table 2
Norms and Reliability Coefficients for the Social Skills Inventory (SSI)

Male subjects Female subjects Total sample

(n=199) (n = 310) (AT =509)
Test-retest
SSI scale M SD M SD M SD (« = 149)

EE 79.8 15.5 83.8 16.7 82.3 16.4 -2.7* .75 .81

ES 90.1 15.3 100.1 13.6 96.2 15.1 -7.7* .78 .90
EC 78.0 14.3 72.8 16.3 74.8 15.7 3.7* .76 .88
SE 84.0 19.4 90.2 19.4 87.8 19.6 -3.5* .88 .96
SS 88.3 15.9 94.2 16.0 91.9 16.2 -4.1* .84 .86
SC 86.5 17.3 87.3 18.9 87.0 18.3 -0.5 .87 .92
SM 72.9 14.4 65.4 16.1 68.3 15.9 5.3* .77 .85
SSI 579.5 65.5 593.8 58.7 588.2 61.8 -2.6' — .94

Note. EE = Emotional Expressivity. ES = Emotional Sensitivity. EC = Emotional Control. SE = Social Expressivity. SS = Social Sensitivity. SC =
Social Control. SM = Social Manipulation.
*p<.05.

skill. For example, a person who is extremely emotionally ex- Factor Analysis of the SSI
pressive may be lacking in certain elements of emotional con-
trol, as the expression of felt emotional states is somewhat spon- In order to provide support for the notion that the seven SSI
taneous and, thus, inconsistent with control. Intercorrelations subscales represent distinct aspects of social skill, a factor analy-
among the seven SSI scales are given in Table 3 separately for sis was conducted with a sample of 339 subjects. Owing to the
male and female subjects. As can be seen, differences in scale length of the SSI, the items for each of the subscales were
intercorrelations for men and women were trivial. Generally, grouped into parcels of five items each, resulting in 21 parcels
the various social skills scales tended to be uncorrelated or of summed items. Care was taken to ensure that each of the
slightly positively correlated with one another. As expected, parcels contained items that represented a cross-section of the
there was a negative correlation between emotional expressivity content of the particular SSI subscale.
and emotional control. An additional negative relation oc- The 21 parcels were factor analyzed using the principal-axis
curred between emotional control and social sensitivity, per- method with an oblique rotation. A seven-factor solution sup-
haps reflecting that individuals high in emotional control tend ported much of factor structure of the SSI. The first factor to
to be self-confident, whereas individuals high in social sensitiv- emerge clearly represented emotional sensitivity. The three par-
ity may be somewhat lacking in self-confidence and may experi- cels containing the emotional sensitivity items had the three
ence some social anxiety. highest factor loadings (.78, .72, .54), no other parcel had a
loading in excess of .34 on this factor.
The second factor to emerge was emotional control. The
Gender Differences three EC parcels had loadings of .49, .59, and .72. The next
highest loadings on this Emotional Control factor were negative
Previous research has found consistent and stable gender
loadings from the EE parcels (-.27 to -.31). These reversed
differences in various communication and social skill dimen-
sions, with women scoring higher on measures of emotional
sensitivity and emotional expressivity (see Friedman et al.,
1980; Hall, 1979; Rosenthal et al., 1979). On the basis of these Table3
results we expected women to score higher than men on the Social Skills Inventory (SSI) Scale Intercorrelations
expressivity and sensitivity scales, and we hypothesized that for Male and Female Subject?
men would score slightly higher on the control and manipula-
SSI scale EE ES EC SE SS SC SM SSI
tion scales. In the present series of studies, women did indeed
score higher on the scales of EE, ES, SE, and SS, whereas men .47 -.18 .57 .19 .47 .17 .68
EE
scored higher on the scales of EC and SM (see Table 2). Through ES .49 — .04 .50 .24 .48 .15 .72
socialization processes and norms for behavior, women become EC -.37 -.05 -.04 -.10 .19 .36 .28
—
SE .49 .40 .01 .20 .62 .11 .77
more skilled in encoding and decoding emotional communica- —
SS .10 .13 -.20 -.16 -.04 .08 .39
tions, whereas men appear to be better able to control their ex- — .72
SC .50 .43 .05 .65 -.33 — .08
pressive behavior and are not as well trained in decoding the SM -.11 .03 .46 -.02 .11 -.11 .45
—
subtleties of emotional expressions. Thus, these gender differ- SSI .62 .67 .25 .73 .15 .68 .37
—
ences in the SSI scales reflect discrepancies that are assumed to
Note. EE = Emotional Expressivity. ES = Emotional Sensitivity. EC =
actually exist. It is important to bear in mind that these gender
Emotional Control. SE = Social Expressivity. SS = Social Sensitivity.
differences are slight and do not necessarily affect the validity SC = Social Control. SM = Social Manipulation.
of the SSI as a research tool. Further exploration of gender " Correlations for male subjects are above the diagonal, female subjects*
differences is beyond the scope of this article. correlations are below the diagonal.
654 RONALD E. RIGGIO

Table 4
Correlations Between Social Skills Inventory (SSI) Scales and Personality

16 Personality Factor Test dimensions

C I
affected by F H tough-
A B feelings- E sober- G shy- minded- L M
reserved- dull- emotionally humble- happy-go- expedient- venture- tender- trusting- practical-
SSI scale outgoing bright stable assertive lucky conscientious some minded suspicious imaginative

EE .33*** .27*** .07 .27*** .53*** -.16* .57*** .09 .06 .00
ES .25" .11 .01 .07 .33*** -.01 ,37... .20* .10 .15
EC -.10 -.05 .16* .00 -.04 .04 -.04 .05 -.03 -.09
SE .45*** .04 .32*** .08 .54*** .01 .61*** .09 -.10 -.05
SS .13 .13 -.40*** -.03 .08 -.06 -.24" .11 .04 -.08
SC .35*** .07 .53*** .06 .37*** .10 .69*** .14 -.02 .08
SM -.16 .08 -.11 .25** .22** -.32*** .12 -.09 .19* .14
SSI .37*** .17* .19* .18* .58*** -.09 .60*** .17* .05 .03

Note. EE = Emotional Expressivity. ES = Emotional Sensitivity. EC = Emotional Control. SE = Social Expressivity. SS = Social Sensitivity. SC :
*p <. 05. **/;<. 01. ***p<. 001.

loadings indicated that emotional expressivity and emotional the SSI scales and traditional trait measures of personality were
control represent, in part, different aspects of a central skill— explored. The correlations between the SSI and these trait mea-
spontaneity versus control of underlying emotions. sures demonstrated the convergent and discriminant validity of
Social sensitivity represented the third factor. Loadings of the the SSI and provided further understanding of the social skill
three SS parcels were .54, .68, and .70. No other loading ex- constructs. The relations between the SSI and other measures
ceeded .25. of communication skill (e.g., the SMS, the ACT, etc.) were also
The fourth factor was social manipulation. Two of the SM explored.
parcels had high factor loadings (.65 and .77), and one loading
was moderate (.25). No other parcel loaded higher than .19 on Method
the Social Manipulation factor.
The fifth factor represented social expressivity with factor One hundred forty-nine undergraduate volunteers (65 men, 84
women) were administered the SSI, the ACT, the SMS (including the
loadings of .54, .63, and .64 for the three SE parcels with no
three SMS subfactors used in Briggs et al., 1980 and Riggio & Friedman,
other parcels loading higher than. 18.
1982), and the short form of the 16 Personality Factor Test (16 PF; Cat-
Social control was clearly the sixth factor to emerge, with the
tell, Eber, & Tatsuoka, 1970), the Private and Public Self-Consciousness
SC parcels all loading above .51 on this factor (.51 to .58). No Scales (Buss, 1980; Fenigstein, Scheier, & Buss, 1975), and the Marlowe-
other parcel's factor loading exceeded .32 on the social control Crowne Social Desirability Scale (SDS; Crowne & Marlowe, 1964).
factor. A subgroup of these subjects (n = 54; 21 men, 33 women) were photo-
The seventh factor did not show any clear pattern of loadings. graphed. A group of untrained students (3 men, 6 women) judged each
None of the parcels loaded near the .50 range that was repre- of the subjects' still photographs on a 9-point scale of physical attractive-
sented in the majority of the significant loadings in the previous ness that ranged from not at all attractive (1) to extremely attractive (9).
six factors. These attractiveness ratings were highly reliable (a = .89) and did not
vary significantly for male or female subjects or for male or female
The one SSI subscale that did not show a distinct factor in
judges. Therefore, the rati ngs were combined into a single index of phys-
the factor analysis was emotional expressivity. Although the EE
ical attractiveness for subjects in this subgroup.
parcels had moderate reversed loadings on the Emotional Con- A group of 119 undergraduate volunteers (44 men, 75 women) were
trol factor (—.27 to —.31) there were also some small loadings administered the second (prefinal) version of the SSI and the PONS
on the Emotional Sensitivity factor (.10 to .24), the Social Ex- (Rosenthal et al., 1979).
pressivity factor (. 11 to . 16), and the Social Control factor (.19
to .32). This appears to represent that the various social skill
Results and Discussion
dimensions are somewhat intercorrelated. The EE scale shared
some variance with several of the other SSI scales. Social skills and the 16 PF. Correlations among the seven
In sum, the results of the factor analysis showed strong sup- scales of the SSI and the 16 PF test are presented in Table 4.
port for the notion that at least six of the seven SSI subscales Individuals scoring high on the EE scale have been described
represent distinct constructs. These results were encouraging as outgoing, emotionally charged individuals. They are people-
and suggested the utility of the SSI as a tool for assessing distinct oriented and active. Correlations between the EE scale and the
areas of basic social skills. 16 PF scales substantiated these descriptions and provided a
richer definition of the construct. Persons who scored high on
Social Skills and Personality: Convergent and the EE scale can be described, in 16 PF terms, as being more
Discriminant Validity outgoing, bright, assertive, happy-go-lucky, expedient, venture-
In order to demonstrate that the SSI scales validly assess im- some, apprehensive, group dependent, and tense. Each of these
portant characteristics of the individual, the relations between relations was consistent with our definition of emotional ex-
 BASIC SOCIAL SKILLS 655

O
self- Q2 Q3 Public Private
N assured- Ql group- careless of Q4 Marlowe- Self- Self- Physical
forthright- apprehen- conservative- dependent- social rules- relaxed- Crowne Conscious- Conscious- Social Attractive-
astute sive liberal self-sufficient controlled tense SDS ness ness Anxiety ness

-.08 .18* -.05 -.23** -.15 .26** -.15 .12 .12 -.26*** .08
-.07 .19* .25** -.12 -.08 .19* .12 .18* .34*** -.21** .09
-.07 -.17* .19* -.02 .20* -.41*** .10 -.15 -.10 -.23** -.07
-.09 .01 .02 -.42*** .03 -.04 .26" .09 .03 -.58*** .13
.41**' .66*** -.30*** -.12 -.20* .57*** -.31*** .58*** .15 .38*** .07
-.30*** -.30*** .27*** -.23** .16 -.27*** .48*** -.26** .04 -.78*** .07
.06 .20* .12 .07 -.26*** .12 -.50*** .07 .17* -.06 -.09
-.05 .18* .14 -.32*** -.06 .09 .04 .16* .19* -.52*** .09

Social Control. SM = Social Manipulation. SDS = Social Desirability Scale. All № = 149, except for physical attractiveness where n = 49.

pressivity with the exception of the correlations with apprehen- Social control was significantly correlated with being outgo-
siveness and tenseness. However, the tenseness or anxiousness ing, emotionally stable, happy-go-lucky, venturesome, forth-
of emotionally expressive individuals may have reflected their right, self-assured, liberal, group dependent, and relaxed. These
high level of felt emotions and a concern with communicating relations support the notion that the individual who scores high
these felt emotions to others. on the SC scale actively takes part in social interactions and is
Individuals who received high scores on the ES scale of the highly knowledgeable and confident in these social encounters.
SSI tended to score higher on several 16 PF dimensions. These Finally, persons who scored high on the SM scale tended to
emotionally sensitive individuals were also more outgoing, hap- score higher on the 16 PF dimensions that describe them as
py-go-lucky, venturesome, tender-minded (sensitive), appre- being assertive, happy-go-lucky, expedient, suspicious, appre-
hensive, liberal, and tense than were individuals who scored low hensive, and careless of social rules. High social manipulators
on the ES scale. The correlations between emotional sensitivity are individuals who are willing to exceed the boundaries of so-
and apprehensiveness, tenseness, and tender-mindedness were cial dictates and bend or break social rules if need be. Because
expected. The correlations with the venturesome, outgoing, and of their nonconventional orientation they may be cautious and
happy-go-lucky scales were not expected and may reflect the wary of others (i.e., they may perceive others to be, like them-
tendency for emotional expressivity and emotional sensitivity, selves, social manipulators). The pattern of correlations with
as we have presented them, to be somewhat positively intercor- the 16 PF factors was consistent with the definition of social
related (i.e., individuals who are skilled at sending emotions manipulation.
may also be slightly more able to detect emotions in others). Social skills and physical attractiveness. Two decades of re-
Emotional control was correlated with the emotional stabil- search have shown that physically attractive persons have a dis-
ity, self-assuredness, liberal, controlled, and relaxed scales of tinct advantage in interpersonal encounters because they tend
the 16 PF, which is consistent with the self-confident, self-con- to make more favorable impressions on others (see Berscheid &
trolled image of the individual who scores high on the EC scale. Walster, 1978). Socially skilled individuals should also have
Social expressivity, like emotional expressivity, was signifi- such an advantage. However, it was hypothesized that the pos-
cantly positively correlated with the 16 PF dimensions that de- session of social skills is independent of physical attractiveness
scribe individuals as outgoing, happy-go-lucky, venturesome, because social skills involve learned abilities. Thus, one test of
and group dependent. These relations reflect the gregarious, ex- discriminant validity involved the relations between the SSI
hibitive nature of the person who scores high on the social ex- scales and the ratings of physical attractiveness. There were no
pressivity dimension. In addition, social expressivity was posi- significant correlations between any of the SSI scales and the
tively correlated with emotional stability, which indicates, per- physical attractiveness index (see Table 4).
haps, the independence of the constructs of EE and SE. That is, Social skills and social desirability. Because it seemed that
the effusiveness of the socially expressive person is of a verbal the possession of high levels of social skills was socially desirable
or social nature rather than the expression of felt emotions char- (i.e., socially skilled persons are evaluated positively and are
acteristic of the emotionally expressive individual. successful in social interactions), a critical test of the SSI's abil-
Individuals who scored high on the SS scale were also high ity to accurately assess underlying social skills involved the
on the "affected by feelings," shyness, astuteness, apprehensive- scales' relation to the Marlowe-Crowne Social Desirability
ness, conservative, tense, and undisciplined self-conflict dimen- Scale (Crowne & Marlowe, 1964). None of the three emotional/
sions of the 16 PF. All these relations were consistent with the nonverbal skill scales were significantly correlated with the SDS
definition of SS, with the exception of the correlations with un- (see Table 4). However, of the four social skill scales, SE and SC
disciplined self-conflict, which was opposite of the expected di- were significantly positively correlated with social desirability,
rection. and SS and SM were significantly negatively correlated with the
656 RONALD E. RIGGIO

SDS. Thus, nonverbal skill seems unrelated to social desirabil- Table 5

ity whereas social or verbal skills do contain some elements of Correlations Between Social Skills Inventory (SSI) Scales and
social desirability. Because many of the social skills items ask Measures ofNonverbal Social Skills
questions about social tactfulness and behavior in interpersonal
situations, these correlations were not surprising. It was sup- SMS factors
posed that these relations between the SDS and the social skills SSI Extra- Other-
scales of the SSI reflect some shared method or item content scale version Acting Directedness Total ACT PONS
variance.
Social skills, self-consciousness, and social anxiety. Buss and EE .41*** .26*** -.05 .29*** .67*** .19*
his colleagues (Buss, 1980; Fenigstein, Scheier, & Buss, 1975) ES .34*** .22** -.07 .19* .53*** .18*
EC .17* .45*** .20* .34*** -.01 -.04
have theorized that there are two types of self awareness, or self- SE .60*** .32*** -.17* .29*** .72*** .05
consciousness—private self-consciousness and public self-con- SS -.21** -.19* .36*** .08 -.08 .10
sciousness. Private self-consciousness deals with awareness of SC .70*** .46*** -.37*** .21** .71*** .11
internal states available only to the individual, such as feelings, SM .17* .47*** .46*** .58*** .20** -.20*
SSI .63*** .55*** .07 .53*** .78*** .12
thoughts, and internal body states. Public self-consciousness
concerns awareness of aspects of the self that are easily observed Note. EE = Emotional Expressivity. ES = Emotional Sensitivity. EC =
by others—appearance, nonverbal cues, and social behaviors. Emotional Control. SE = Social Expressivity. SS = Social Sensitivity.
As much of private self-consciousness deals with awareness of SC = Social Control. SM = Social Manipulation. SMS = Self-Monitor-
felt emotional states it seems theoretically linked to emotional ing Scale. ACT = Affective Communication Test. PONS = Profile of
Nonverbal Sensitivity. All Ns = 149, except for the PONS where n =
sensitivity, because sensitivity to personal emotional states is 119.
likely a prerequisite for skill in detecting emotional states in *p <.05. **p<.01. ***/»<.001.
others. Public self-consciousness, on the other hand, should be
related to the skill of social sensitivity, because it deals with con-
cern over social displays. Correlations between the SSI scales
and the Private and Public Self-Consciousness scales were con- nipulation. However, all of these relations were somewhat weak.
sistent with these hypotheses. Private self-consciousness was sig- The negative relation between PONS and social manipulation
nificantly positively correlated with emotional sensitivity and may reflect some lack of concern for, or lack of attention to,
public self-consciousness was strongly positively correlated the feelings of others in the high social manipulator. No other
with social sensitivity (see Table 4). These patterns of corre- correlations between SSI scales and PONS reached significance.
lations provide further validation for the SSI, particularly for Although the ACT was designed to measure emotional ex-
the two sensitivity scales. pressivity, some of the items appear to assess dimensions of so-
An additional dimension proposed by Buss, his colleagues, cial expressivity and social and emotional control. Consistent
and others (Leary, 1983), is the trait of social anxiety. Social with this notion, the ACT showed strong positive correlations
anxiety is defined as "discomfort in the presence of others" with the EE, SE, and SC scales of the SSI. Contrary to predic-
(Buss, 1980, p. 204). Social anxiety is strongly related to, and tion, the ACT was uncorrelated with emotional control, but it
perhaps develops from, a lack of social skills, particularly skill was positively related to emotional sensitivity. Thus, the ACT
in expressivity and control. Consistent with this notion, the so- may assess a broader range of emotional and social communica-
cial anxiety scale was significantly negatively correlated with tion skills than was previously believed.
the three emotional skill scales, social expressivity, and social Inasmuch as the definition of the construct of self-monitoring
control (see Table 4). However, social anxiety was significantly seems to include a wide array of basic social skills, the patterns
positively correlated with social sensitivity, perhaps reflecting of relations between the SSI scales and the SMS factors are criti-
the socially sensitive individual's strong concern with the appro- cal to the validation of the SSI and for the further understanding
priateness of social behavior. of individual differences in social skills. The three SMS factors
Social skills, self-monitoring, and nonverbal skill. As men- outlined in Briggs et al. (1980) seem to represent the three basic
tioned earlier, the basic social skill dimensions were closely re- dimensions of communication skill, although there may be con-
lated, in predictable patterns, to previous research that exam- siderable overlap among these factors (i.e., they are not orthogo-
ined basic nonverbal skill constructs, particularly the research nal). Also, they are not highly reliable because they contain few
of Rosenthal and his colleagues (Rosenthal et al., 1979) on non- items.
verbal sensitivity, the work of Friedman and his colleagues SMS-Extraversion would seem to be most closely related to
(Friedman et al., 1980) on nonverbal expressiveness (also called expressivity, both social and emotional; but it may also contain
personal charisma), and especially Snyder's (1974, 1979) work some elements of control, because some of the SMS-Extraver-
with self-monitoring. sion items deal with social control or acting. In our sample,
The correlations between the SSI scales and the PONS, ACT, SMS-Extraversion showed a significant positive relation to all
SMS, and the SMS factors of Extraversion, Acting, and Other- of the SSI scales, except for social sensitivity with which it was
Directedness are presented in Table 5. It was predicted that the significantly negatively correlated.
total PONS score, a performance index of nonverbal sensitivity, SMS-Acting should be most closely related to dimensions of
would be positively related to SSI sensitivity, particularly emo- control and social manipulation, but it also includes some ele-
tional sensitivity. The PONS was significantly positively corre- ments of expressivity. Like SMS-Extraversion, SMS-Acting
lated with emotional sensitivity and was also positively related was positively correlated with all of the SSI scales, except for a
to emotional expressivity and negatively related to social ma- negative correlation with SS. The strongest correlations, how-
 BASIC SOCIAL SKILLS 657

ever, were between SMS-Acting and the EC, SC, and SM scales The correlations between the SSI and the various indexes of
of the SSI. social roles and social behaviors are presented in Table 6.
Finally, SMS-Other-Directedness is theoretically related to Contrary to expectations, present employment, expected fu-
the skills of social sensitivity and social manipulation. Both of ture employment, and ideal employment were not consistently
these correlations were strongly positive in our sample. Other- significantly related to the SSI scales, although correlations be-
Directedness was also slightly positively correlated with emo- tween total SSI and the three employment indexes approached
tional control and was negatively related to social expressivity significance for future and ideal employment. However, subjects
and social control. who were presently employed did tend to be more socially
The patterns of correlations between the SSI scales and the skilled (on the basis of total SSI) than were unemployed sub-
other, preexisting measures of social skills provide further vali- jects, r(60) = .30, p < .05. This lack of expected findings may
dation for the SSI scales as denned. The SSI does indeed appear be based, in part, on the fact that the undergraduate subjects,
to assess basic nonverbal and social skill dimensions in relation who were mostly part-time employees, were not currently em-
to similar self-report skill measures and the performance mea- ployed in positions that best fit their social skills. Also, their
sure of nonverbal sensitivity. expectations about future careers may not take into consider-
ation the social abilities they possessed that would enable them
Social Skills, Social Roles, and Social Behaviors to perform the specific job well.
Additional research has found that total scores on the SSI
Because basic social skills should play an important role in predicted students' performance in mock hiring interviews,
governing complex social behaviors, it seemed reasonable to as- perhaps a more appropriate test of the validity of the SSI when
sume that individual differences in basic social skill dimensions using student populations (Throckmorton, 1985).
were related to persons' social group memberships, the social Individuals who scored high on the scales of SE, SC, and total
roles they adopt, and the particular types of social behaviors SSI did, as expected, tend to have more sales experience than
they perform. In particular, it was assumed that socially skilled did persons scoring low on SE, SC, and SSI. The verbal fluency
individuals would (a) seek out employment (and be employed) and other-oriented behaviors characteristic of socially expres-
in occupations that allowed them more social contacts and sive persons should contribute to their aptitude for and success
greater use of social skills; (b) have more experience in theatri- in sales, whereas the skill of social control may enhance verbal
cal acting and lecturing than nonsocially skilled persons; (c) be persuasion.
more likely to hold an office in a club or organization; and (d) Persons who scored high on EC, a key component of theatri-
have more social ties than their nonskilled counterparts (see cal acting, tended to have more acting experience than persons
Friedman et al., 1980). scoring low on EC. Similarly more socially skilled persons, as
evidenced by high total SSI scores, were more likely to have had
Method acting experience and taken acting classes than were individuals
who scored low on the SSI.
Sixty undergraduate volunteers (24 men, 36 women) were adminis- Contrary to prediction, lecturing experience and being
tered the SSI and a self-report questionnaire that asked about the follow-
elected to a political office in a club or organization were not
ing: present, future, and ideal employment, past sales experience, acting
significantly correlated with the SSI. However, persons who
experience, acting classes, lecturing experience, dating activity, number
scored high on total SSI and the scales of EC, SC, and SM re-
of close friends, number of school acquaintances, elected offices held in
ported that they would feel more comfortable lecturing to a
clubs and organizations, a 9-point rating scale of how comfortable (very
uncomfortable to very comfortable) they would be "giving a presentation large group of people than did persons scoring low on these di-
in front of a large class," and a 9-point self-rating of shyness (not at all mensions. The lecturing and political experiences of these stu-
shy or extremely shy). dent subjects may have been more limited and restricted than
the general population, thus leading to the lack of expected
findings. Also, lecturing may be governed more by the school
Results and Discussion
environment (i.e., course requirements) than by subjects' social
In order to determine the extent to which certain occupations skills.
required the use of social skills, a group of judges rated each of As predicted, socially skilled persons (on the basis of total
the present, future (expected), and ideal jobs on four social skill SSI, and high EE, SE, ES, and SC scores) reported more social
related scales. The 5-point scales rated each job in terms of (a) ties than did nonskilled subjects, both in terms of close friends
the amount of contact with other people required by the job; and number of daily school acquaitances. Total SSI has also
(b) the amount of expressivity required by the job; (c) the been shown to be significantly positively correlated with stan-
amount of sensitivity the job required; and (d) the amount of dardized measures of social support and social network size
control the job required. The same 5 judges (1 male, 4 female) (Zimmerman & Riggio, 1986). There were no significant re-
made each of the four ratings independently, on separate occa- lations between dating frequency and SSI scales.
sions. All of the judges were thoroughly familiar with the defi- Finally, socially skilled individuals and persons scoring high
nitions of the social skill dimensions. All of the ratings were on EE, SE, and SC rated themselves as less shy than did persons
highly reliable (a = .83 to .94). Because the four ratings for each scoring low on these dimensions. Subjects high on SS had a
set of jobs (i.e., present, future, ideal) were highly intercorre- slight tendency to rate themselves as more shy than did persons
lated (rs = .66 to .94), the four scales were summed, which re- low on SS.
sulted in a general index reflecting the level of general social In sum, these results indicate some consistent relations be-
skills each job required. tween the possession of various social skills, as assessed by the
658 RONALD E. RIGGIO

Table 6
Correlations Between Social Skills, Social Roles, and Self-Reported Social Behaviors

SSI scales

Behavior and roles EE ES EC SE SS SC SM SSI

Present employment 42 .05 .04 .07 -.06 .15 -.15 .15 .05
Future employment 56 .19 .14 .04 .17 .08 .25* -.02 .26*
Ideal employment 57 .15 .15 -.10 .26** .02 .30** -.12 .22*
Sales experience 60 .18 .20 -.10 .35*** -.17 .28** .01 .25*
Acting experience 60 -.01 .12 .24* .13 .05 .09 .13 .22*
Acting classes 60 .13 .11 .08 .26** -.02 .14 -.02 .22*
Lecturing experience 59 -.06 -.05 .06 .02 -.01 .23* -.03 .05
Number of dates 23 .08 .23 -.14 .13 .11 -.10 -.16 .04
Number of close friends 59 .36*** .37*** .03 .24* .19 .34*** .16 .49***
Number of daily acquaintances 57 .27** .28** .05 .38*** -.20 .45*** .02 .40***
Elected offices 59 -.02 .13 .09 -.03 -.05 .06 -.18 .00
Speaking comfort 60 .19 -.02 .29** .20 -.18 .42**« .25* .36***
Shyness 60 -.44*** -.25* .03 -.58*** .24* -.59*** .00 -.53***

Note. SSI = Social Skills Inventory. EE = Emotional Expressivity. ES = Emotional Sensitivity. EC = Emotional Control. SE = Social Expressivity.
SS = Social Sensitivity. SC = Social Control. SM = Social Manipulation.

SSI scales and self-reports of social behaviors—important evi- this the conversation was allowed to take its own course, led by the sub-
dence for the validity of the SSI. ject. After 2 min the experimenter interrupted the conversation and
informed the subject that the study was over. Before the subject was
dismissed, the experimenter took a full-body photograph of the subject.
Social Skill and Initial Attraction Immediately after the subject departed, the confederates and experi-
If possession of certain basic social skills is indeed related to menter rated the subject on a 9-point scale of initial likability (scale
anchors: not at all likable to extremely likable). The reliability coeffi-
social effectiveness we would expect socially skilled individuals
cient for these three ratings was acceptable (a = .77). Therefore, the
to have greater success in social interactions than would non-
ratings were summed to form a single index of initial likability.
skilled persons. A wealth of research on person perception indi-
Throughout each interaction the confederates and experimenter were
cates the tremendous importance of initial impressions in instructed to keep their behavior as standard as possible. All partici-
affecting the outcome of subsequent social interaction (Schnei- pants were blind as to subjects' SSI scores, and the confederates were
der, Hastorf, & Ellsworth, 1979). An individual's ability to unaware of the purposes of the study.
make a good initial impression can be a critical determinant A group of 12 student volunteers rated subjects' photographs on a 9-
of future social success. By our definitions, social expressivity, point scale of physical attractiveness. These ratings were averaged into
emotional expressivity, and social control are the three basic a single, reliable (a = .92) index.
skills that should be most closely related to the formation of
initial impressions. Highly expressive and socially controlled Results and Discussion
persons should receive favorable evaluations from others in ini-
tial encounters. However, as mentioned earlier, physical attrac- As can be seen in Table 7, the SE, SC, and total SSI scores
tiveness has also been shown to be a consistent predictor of suc- were significantly correlated with ratings of initial likability. Be-
cess in initial interaction (Berscheid & Walster, 1978). Our ear- cause expressivity was thought to be a key determinant of initial
lier investigations showed physical attractiveness and social attraction, the EE and SE scales were summed to form a single
skills to be relatively independent constructs. If this is true we index. Scores on this expressivity index were also positively re-
would expect possession of both physical attractiveness and so- lated to likability ratings.
cial skills to be positively related to impressions that individuals Surprisingly, physical attractiveness was significantly posi-
make in initial encounters. tively correlated with total SSI in this group of subjects, r(6l) =
.30, p < .05—in direct contrast to our earlier findings. Because
of this, it was important to examine the effects expressivity and
Method
social skills had on ratings of initial likability independent of
Sixty-one students (25 men, 36 women) volunteered for a study of the effects of physical attractiveness. Therefore, Table 7 also
role-playing. Upon arrival at the laboratory, each subject was greeted by presents the partial correlations of the summed expressivity
the experimenter, led to a room, and administered the SSI. Each subject
scales and total SSI with initial likability, partialing out the
was then told that he or she would be meeting two other subjects with
effects of physical attractiveness. The relations were essentially
whom the subject would be engaging in some role-playing. The subject
unchanged. Possession of social skills, particularly expressivity
and experimenter then entered the adjoining room where two confeder-
ates (one male, one female) sat facing an empty chair. The experimenter and SC, did indeed seem to predict the favorability of initial
instructed the subject to "go over and introduce yourself to these two impressions. These results hint at the importance of social skills
people and take a seat" The confederates asked each subject the same in affecting social outcomes as well as in providing evidence of
questions about college major and role-playing experiences. Aside from the predictive validity of the SSI.
 BASIC SOCIAL SKILLS 659

Table 7 sensitivity; see Johnson, Cheek, & Smither, 1983). The Empa-
Relation Between Social Skills, Physical Attractiveness, thy Scale may be a general index of social skills. Charisma, al-
and Ratings of Initial Likability though dominated by skill in emotional expressivity (see Fried-
man et al., 1980), may actually reflect global, balanced posses-
Partial correlations:
sion of all the basic dimensions of social skill. Thus, individuals
Initial likability
Initial controlling for who are labeled charismatic may simply be persons who have
Scale likability physical attractiveness extraordinary skills in dealing with others.
Since basic social skills represent learned communication
EE .19
abilities, it seems reasonable that people can develop and en-
SE .32**
SC .39** hance these basic skills. Practice and training in basic dimen-
EE&SE .31** .30** sions of social skills and in the interrelationships of the basic
SSI .25** .20* skills should lead to more effective social performance.
Physical Attractiveness .20 Although the SSI was designed primarily as a research tool—
a means of assessing individual differences in social communi-
Note. SSI = Social Skills Inventory. EE = Emotional Expressivity. SE =
Social Expressivity. SC = Social Control. All «s = 61. cation skills for use in studying social interaction—it may even-
*/><.06.**p<.05. tually have many practical applications in industrial and organ-
izational settings, education, and therapy. The SSI can be used
in conjunction with performance measures of social skill or in
instances where performance measures are impractical.
General Discussion
Although this article outlines some of the possible basic di-
Although there have been several attempts to measure a spe- mensions of communication skill, much additional research is
cific dimension of nonverbal skill (e.g., Friedman et al, 1980; needed to further understand the basic components that com-
Rosenthal et al., 1979) and, more recently, attempts to assess prise social skill.
social skill-related constructs such as empathy, shyness, and so-
ciability (e.g., Cheek & Buss, 1981; Greif & Hogan, 1973; Ho-
gan, 1969), there has been no integrated approach to assessing
References
multiple dimensions of social skill. The SSI represents an initial Archer, D., & Akert, R. M. (1977). Words and everything else: Verbal
attempt to provide a general framework for nonverbal and so- and nonverbal cues in social interpretation. Journal of Personality
cial skill research. and Social Psychology, 35, 443-449.
The results of the series of initial validation studies for the Berscheid, E., & Walster, E. H. (1978). Interpersonal attraction (2nd
SSI were encouraging. Patterns of correlations among the SSI ed.). Reading, MA: Addison-Wesley.
scales and traditional measures of personality and other indexes Briggs, S. R., Cheek, J. M., & Buss, A. H. (1980). An analysis of the
of social skills demonstrated the convergent and discriminant Self-Monitoring Scale. Journal of Personality and Social Psychology,
38,679-685.
validity of the SSI. Moreover, correlations among the SSI scales
Buck, R. (1976). A test of nonverbal receiving ability: Preliminary stud-
and self-reported performance of social behaviors, such as act-
ies. Human Communication Research, 2,162-171.
ing and sales experience and number of social ties, provided
Buck, R. (1978). The slide-viewing technique for measuring nonverbal
evidence of the concurrent validity of the SSI. These subjects sending accuracy: A guide for replication. JSAS Catalog of Selected
sought social group memberships and exhibited social behav- Documents in Psychology, 63.
iors in patterns that could be predicted from their SSI scores. Buck, R. (1983). Nonverbal receiving ability. In J. M. Wiemann & R.
As expected, individuals who were more expressive and so- P. Harrison (Eds.), Nonverbal interaction (pp. 209-242). Beverly
cially skilled were evaluated more positively by observers in ini- Hills, CA: Sage.
tial encounters. This last finding indicates that the basic skill Buck, R. (1984). The communication of emotion. New York: Guilford
approach may further our understanding of the social interac- Press.
Buss, A. H. (1980). Self-consciousness and social anxiety. San Fran-
tion process and help to uncover some of the basic elements that
cisco: Freeman.
lead to success in interpersonal encounters.
Cattell, R. B., Eber, H. W., & Tatsuoka, M. M. (1970). Handbook for the
By using the social skill approach to break a complex behav- 16 PF(4th ed.). Champaign, IL: Institute for Personality and Ability
ior pattern into its more basic skill components, we may begin Testing.
to understand the dynamics of what are now poorly understood Cheek, J. M., & Buss, A. H. (1981). Shyness and sociability. Journal of
social phenomena. For example, psychologists and laypersons Personality and Social Psychology, 41, 330-339.
alike have been intrigued by elusive concepts such as empathy, Christie, R., & Geis, F. (1970). Studies in Machiavellianism. New York:
intuition, and charisma. It may be the case that individuals who Academic Press.
are labeled empathic, intuitive, or charismatic possess certain Crowne, D. P., & Marlowe, D. (1964). The approval motive. New York:
combinations of basic social skills that make up these character- Wiley.
Curran, J. P., & Monti, P. M. (Eds.). (1982). Social skills training. New
istics. Intuition may be traced to sensivitity, particularly to skill
York: Guilford Press.
in decoding subtle emotional communications. Empathy may Davis, M. H. (1983). Measuring individual differences in empathy: Evi-
be composed of skill in emotional sensitivity paired with emo- dence for a multidimensional approach. Journal of Personality and
tional and social expressivity (i.e., being sensitive to the emo- Social Psychology, 44, 113-126.
tions of another and ability to express empathic concern). In- DePaulo, B. M., & Rosenthal, R. (1979). Telling lies. Journal of Person-
deed, recent work with Hogan's (1969) Empathy Scale may in- ality and Social Psychology, 37, 1713-1722.
dicate the presence of more basic, skill-related factors (e.g., Diez, M. E. (1984). Communicative competence: An interactive ap-
660 RONALD E. RIGGIO

proach. In R. N. Bostrom (Ed.), Communication yearbook 8 (pp. 56- to deception. Journal of Personality and Social Psychology, 45, 899-
79). Beverly Hills, CA: Sage. 915.
Dymond, R. F. (1949). A scale for the measurement of empathic ability. Riggio, R. E., Widaman, K. F., & Friedman, H. S. (1985). Actual and
Journal of Consulting Psychology, 13,127-133. perceived emotional sending and personality correlates. Journal of
Fenigstein, A., Scheier, M. F., & Buss, A. H. (1975). Public and private Nonverbal Behavior, 9, 69-83.
self-consciousness: Assessment and theory. Journal of Consulting and Rosenthal, R. (Ed.). (1979). Skill in nonverbal communication. Cam-
Clinical Psychology. 43, 522-527. bridge, MA: Oelgeschlager, Gunn, & Hain.
Friedman, H. S. (1979). The concept of skill in nonverbal communica- Rosenthal, R., Hall, i. A., DiMatteo, M. R., Rogers, P. L., & Archer,
tion: Implications for understanding social interaction. In R. Rosen- D. (1979). Sensitivity to nonverbal communication: The PONS test.
thai (Ed.), Skill in nonverbal communication (pp. 2-27). Cambridge, Baltimore, MD: Johns Hopkins University Press.
MA: Oelgeschlager, Gunn, & Hain. Schneider, D. J., Hastorf, A. H., & Ellsworth, P. C. (1979). Person per-
Friedman, H. S., Prince, L. M., Riggio, R. E., & DiMatteo, M. R. ception. Reading, MA: Addison-Wesley.
(1980). Understanding and assessing nonverbal expressiveness: The Snyder, M. (1974). The self-monitoring of expressive behavior. Journal
Affective Communication Test. Journal of Personality and Social Psy- of Personality and Social Psychology, 30, 526-537.
chology, .?9, 333-351. Snyder, M. (1979). Self-monitoring processes. In L. Berkowitz (Ed.),
Friedman, H. S., & Riggio, R. E. (1981). Effects of individual differences Advances in experimental social psychology (Vol. 12, pp. 85-128).
in nonverbal expressiveness on transmission of emotion. Journal of New York: Academic Press.
Nonverbal Behavior, 6, 96-102. Thorndike, E. L. (1920). Intelligence and its uses. Harper's Magazine,
Galassi, J. P., Galassi, M. D., & Vedder, M. J. (1981). Perspectives on 140, 227-235.
assertion as a social skills model. In J. D. Wine & M. D. Smye (Eds.), Thorndike, R. L. (1936). Factor analysis of social and abstract intelli-
Social Competence (pp. 287-345). New York: Guilford Press. gence. Journal of Educational Psychology, 27,231-233.
Greif, E. G., & Hogan, R. (1973). The theory and measurement of em- Thorndike, R. L., & Stein, S. (1937). An evaluation of the attempts-to
pathy. Journal of Counseling Psychology, 20, 280-284. measure social intelligence. Psychological Bulletin, 34, 275-285.
Guilford, J. P. (1967). The nature of human intelligence. New York: Throckmorton, B. (1985). Interrelationships of communication skills
McGraw-Hill. and interviewing success. Unpublished master's thesis, California
Hall, J. A. (1979). Gender, gender roles, and nonverbal communication State University, Fullerton.
skills. In R. Rosenthal (Ed.), Skill in nonverbal communication (pp. Trower, P., Bryant, B., & Argyle, M. (1978). Social skills and mental
32-67). Cambridge, MA: Oelgeschlager, Gunn, & Hain. health. Pittsburgh, PA: University of Pittsburgh Press.
Hogan, R. (1969). Development of an empathy scale. Journal of Con- Watson, D., & Friend, R. (1969). Measurement of social-evaluative anx-
sulting and Clincial Psychology, 33, 307-316. iety. Journal of Personality and Social Psychology, 33, 448-457.
Hunt, T. (1928). The measurement of social intelligence. Journal of Ap- Wiemann, J. M. (1977). Explication and test of a model of communica-
plied Psychology, 12, 317-334. tive competence. Human Communication Research, 3,195-213.
Johnson, J. A., Cheek, J. M., Smither, R. (1983). The structure of empa- Wiemann, J. M., & Backlund, P. (1980). Current theory and research
thy. Joumalof Personality and Social Psychology, 45, 1299-1312. in communicative competence. Review of Educational Research, 5,
Leary, M. (1983). Understanding social anxiety. Beverly Hills, CA: 185-199.
Sage. Wine, J. D., & Smye, M. D. (1981). Social competence. New York: Guil-
Lennox, R. D., & Wolfe, R. N. (1984). Revision of the Self-Monitoring ford Press.
Scale. Journal of Personality and Social Psychology, 46, 1349-1364. Zimmerman, J. A., & Riggio, R. E. (1986). Social skills and social sup-
McCroskey, J. C. (1977). Oral communication apprehension: A sum- port. Unpublished manuscript.
mary of recent theory and research. Human Communication Re- Zuckerman, M., Hall, J. A., DeFrank, R. S., & Rosenthal, R. (1976).
search, 4, 78-96. Encoding and decoding of spontaneous and posed facial expressions.
Mehrabian, A., & Epstein, N. A. (1972). A measure of emotional empa- Journal of Personality and Social Psychology, 34, 966-977.
thy. Journal of Personality, 40, 525-543. Zuckerman, M., & Larrance, D. (1979). Individual differences in per-
Meichenbaum, D., Butler, L., & Gruson, L. (1981). Toward a concep- ceived encoding and decoding abilities. In R. Rosenthal (Ed.), Skill
tual model of social competence. In J. D. Wine & M. D. Smye (Eds.), in nonverbal communication (pp. 171-203). Cambridge, MA:
Social competence (yp- 36-60). New York: Guilford Press. Oelegeschlager, Gunn, & Hain.
O'Sullivan, M. (1983). Measuring individual differences. In J. M. Wie- Zuckerman, M., Larrance, D., Hall, J., DeFrank, R., & Rosenthal, R.
mann & R. P. Harrison (Eds.), Nonverbal interaction (pp. 243-270). (1979). Posed and spontaneous communication via facial and vocal
Beverly Hills, CA: Sage. cues. Journal of Personality, 47, 712-733.
Rathus, S. A. (1973). A 30-item schedule for assessing assertive behav- Zuckerman, M., Lipets, M. S., Koivumaki, J. H., & Rosenthal, R.
ior. Behavior Therapy, 4, 398-406. (1975). Encoding and decoding nonverbal cues of emotion. Journal
Riggio, R. E., & Friedman, H. S. (1982). The interrelationships of self- of Personality and Social Psychology, 32, 1068-1076.
monitoring factors, personality traits, and nonverbal social skills.
Journal of Nonverbal Behavior, 7, 33-45. Received July 26, 1985
Riggio, R. E., & Friedman, H. S. (1983). Individual differences and cues Revision received January 21, 1986 •

View publication stats