EuropeanJournalofPsychologicalAssessment,Vol.15,Issue1,pp.

2538

Assessing Social Anxiety:

The Inventory of
Interpersonal Situations (IIS)
Rien van Dam-Baggen1 and Floris Kraaimaat2
1

Department of Clinical Psychology, University of Amsterdam, The Netherlands

2
Department of Medical Psychology, University of Nijmegen, The Netherlands
Keywords: Questionnaire, social anxiety

Summary:Thepurposeofthisstudywastodevelopaselfreportquestionnairefortheassessmentofsocial
anxiety in adults. The Inventory of Interpersonal Situations (IIS) consists of 35 items formulated as
responsesto specificsocialsituations.TheIISis basedonaninteractiveconceptofsocialanxiety and
providesscoresforbothaDiscomfortandaFrequencyscale.ThereliabilityandvalidityoftheIISwere
investigated in several adult psychiatric and nonpsychiatric samples. The scales for Discomfort and
Frequencyshowedstabilityovertime.Cronbachssrevealedasufficientlyhighinternalconsistencyon
bothscales,whiletheconceptualstructurewasshowntoberatherinvariantacrosssociallyanxiousand
nonsocially anxious groups. The IIS scales were able to discriminate between socially anxious and
nonsocially anxioussamples,andshowedsignificantrelationshipswithindependentmeasuresofsocial
anxiety.TheIISscalesdemonstratedhighpredictivevalidityforovertbehaviorinsocialsituations.
depression, schizophrenia,
orstuttering),butalsoasa
noninandofitself(e.g.,
rather isolated problematic
The terms social anxiety and nonassertiveness have gen DSMIII and subsequent
behavior by itself. There
erallybeenusedascommonsenseconstructs,withsocial editions: APA, 1980,
are also indications that
anxiety referring mainly to the subjective distress 1987, 1994). As far as
recidivism

and
experienced in social situations, and nonassertiveness other behavior is
rehospitalizationofclinical
referringtoovertbehavioralaspectsofsocialbehavior.Inconcerned,suchasshyness
psychiatric patients is
thebehavioralscientificliterature,avarietyofsynonyms andsocialwithdrawal,they
related to high social
has been used for social anxietyand (non)assertiveness, have been conceived of
anxiety and deficient
such as shyness, social inhibition, interpersonal anxiety, onlyassecondaryphenom
social skills (Zigler &
communication apprehension, embarrassment, social ena to other disorders or
Glick,1986).
inadequacy,interpersonaleffectiveness,socialcompetence, syndromes. The advantage
The lack of a
reticence and selfconsciousness (Van DamBaggen &ofa dimensional approach
comprehensive
theory of
Kraaimaat,1989).Thelackofacomprehensivetheoryofto the concept over
social

anxiety
contrasts
social anxiety is reflected in the rather idiosyncratic and psychiatric classification
withitsclinicalimportance
descriptive definitions presented in the field. The broad suchasfoundwithDSMis
(Van DamBaggen &
rangeofconceptionsofsocialanxietycanbeseeninthe that it leads to a more
Kraaimaat, 1989). Three
literaturewithrespect totheprevalenceofsocial anxiety differentiated picture,
hypotheses have been
and nonassertiveness. For instance, in psychiatry and demonstrating that social
postulated,eachservingas
psychopathology handbooks, social anxiety and itsanxiety can be considered
the basis for commonly
synonyms are scarcely mentioned; only social phobia is an

accompanying
used treatment methods:
consideredandrecognizedasaphenome
phenomenon in several
inhibition by anxiety, the
behavioraldisorders(e.g.,
absenceofor

EJPA15(1),1999Hogrefe

&HuberPublishers

26

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)
situations. Discomfort
has been measured by
ule (Wolpe & Lazarus,
insufficiency in social skills, and debilitating questionnaires such as
1966; Hersen, Bellack,
cognitions.These hypothesesare not mutually theSocialAvoidanceand
Turner,

Williams,
exclusive and reflect, analogous with other Distress Scale (Watson
Harper, & Watts, 1979),
anxieties (Lang, 1971), the three aspects of &Friend,1969),theFear
theRathusAssertiveness
social anxiety: emotional, behavioral, and of Negative Evaluation
Schedule (Rathus, 1973)
cognitive. A complicating issue in the Scale(Watson&Friend,
andtheAssertionInven
assessment of social anxiety is that there are 1969; Leary, 1983),The
tory(Gambrill&Richey,
multiple indicators within each aspect. For Social Anxiety Scale
1975). With the
instance,heartrate,skinconductance,andblood (Willems, Tuenderde
exception of the
pressure transformed to autonomic reactivity Haan,&Defares,1973),
Assertion Inventory,
scores have been used as physiologicalthe Social Anxiety
differentinventorieshave
indicatorsoftheemotionalaspect(e.g.,Beidel, Inventory(Richardson&
tobeusedtomeasurethe
Turner, & Dancu, 1985; Turner, Beidel, & Tasto, 1976), the Social
emotional as well asthe
Phobia and Anxiety
Larkin,1986;Bruch,Gorsky,Collins,
behavioral aspect of
Inventory (Turner,
social anxiety. In
1&
Berger, 1989). Molecular measures
Beidel, Dancu, &
addition,

another
such as response latency (e. g., Pitcher &
Stanley, 1989) and the
disadvantage of the
Meikle, 1980; Romano & Bellack, 1980,
Assertion Inventory
aforementioned
Trower, 1980; Asendorpf, 1988), midlevel
(Gambrill & Richey,
questionnairesisthatthe
measures such as gestures (Monti, Boice,
1975).

Reported
socalled

negative
Fingeret, Zwick, Kolko, Munroe, &
frequency has been
domain

of

social
Grunbergen,1984),andmolarmeasuressuchas
measured by question
responses,

i.

e.,
sit
globalratingsofoverallassertiveness(e.g.,St.
nairessuchastheWolpe
uations in which a
Lawrence,1982;Nelson,Hayes,Felton,&Jar
Lazarus Assertiveness
response of assertively
rett,1985)havebeenusedasindicatorsofthe
Sched
standingupforoneselfis
overt behavioral aspect. Finally, subjective
needed,

is
distress (e. g., Schwartz & Gottmann, 1976),
overrepresented

in

the
cognitive selfstatements (e. g., Clark &
item set, while the so
Arkowitz,1975;Alden&Cappe,1981;Glass,
calledpositivedomainof
Merluzzi,Biever,&Larsen,1981)andselective
assertiveness, namely,
informationprocessing propensities (Hope,
situationsthataredirectly
Rapee, Heimberg, & Dombeck, 1990; Cloitre,
orindirectlyaimedatthe
Heimberg,Holt,
exchange of positive
2&
Liebowitz,1992)havebeenusedas
emotions with other
indicatorsofthecognitiveaspect.
persons,

is
Research has shown that high socially
underrepresented.
anxious persons differ from low socially
Inclinicalpractice,the
anxiouspersonsinsocialsituationsintheir(1)
emotional
aspect has
psychophysiological reactions(e. g.,Turner et
obtained

a rather
al., 1986; Bruch et al., 1989), (2) overt
dominant position in the
behavioral reactions (Bruch, 1981; McFall,
measurement of social
Winnett, Bordewick, & Bornstein, 1982; Van
anxiety.Themainreason
DamBaggen & Kraaimaat, 1987b) and (3)
for this is that
cognitive reactions (Heimberg, Chiauzzi,
apprehension (or the
Becker, & MadrazoPeterson, 1983; LaVome
subjective distress in
Robinson&Calhoun,1984;VanDamBaggen
socialsituations)isoften
&Kraaimaat,1987b).Intheclinicalassessment
reportedasadebilitating
ofsocialanxietyandnonassertivenessinadults
aspect of social
and adolescents, the operationalization of the
functioning. Because of
apprehensionaspecthasfocusedmainlyonthe
their convenience and
degree of discomfort experienced in
efficiency, question
interpersonal situations, while the operational
naires are the most
ization of the behavioral aspect
frequently

used
(nonassertiveness) has focused mainly on the
instruments in the
reported frequency ofbehaviorininterpersonal

assessmentofsocialanxiety.Traditionally,self assertive behavior in

reportinventorieswereconstructedtomeasurea social

situations,
trait or a disposition. Mischel (1968, 1973, namely, from both the
1990) vehemently criticized this type ofpositive and negative
questionnaire for failing to predict specificdomains

of
behavior in specific situations. Although this assertiveness (situation
debate was already surpassed by the view in facets, e. g., Stouthard,
whichsocialanxietywasconsideredacomplex Hoogstraten,&Mellen
responsewiththreeaspects,twolessonshadyet bergh, 1995) and (2)
tobelearned,namely:
two reaction facets (e.

(1) selfreport measures should take into g., Stouthard et al.,

account situational specificity; and (2) the1995) of the construct
items should describe behavior (concreteneed to be measured,
social responses) that must be as situation namely, the discomfort
specific as possible. At the same time, the (emotional aspect) in
inventory should be as brief as possible social situations and
(Angleiter, John, & Lhr, 1986). Designingthe frequency of the

response
anddevelopingacomprehensiveinstrument social
(behavioralaspect).
abletomeasuresocialanxietymeansthat(1)
theitemsshould represent a broad range of The aim of the
EJPA15(1),1999

present paper was to

developandinvestigate
the

psychometric
characteristics of the
Inventory

of
Interpersonal Situations
(IIS)

as

a
comprehensive
instrumenttoassessthe
emotionalaswellasthe
behavioral aspect of
social anxiety on
positive and negative
domains

of
assertiveness in clinical
and nonclinical pop
ulations.Severalstudies
were successively
carried out, and are
reportedbelow.

Hogrefe&HuberPublishers

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)
magnitude

of
discomfort experienced
Study 1: Development of the while performing the
social responses de
IIS
scribedintheitemsof
the Discomfort scale,
ThepurposeofStudy1wastodevelopashort,
and (2) the reported
focused, and contentvalid Inventory of
frequency

of
Interpersonal Situations (IIS) to be used for
performingthesesocial
clinicalandresearchpurposes.Theitemshadto
responses in the
reflect the interactive conception of social
Frequency scale. Each
anxiety,whilethequestionnairehadtocomprise
scale has both a
scales for discomfort and frequency. The
separateinstructionand
following criteria were established: (1) to be
administrationform.In
applicableacrossvariouspopulationsand(2)to
the Discomfort scale,
predictspecificbehaviorinsocialsituations.
theitemsarescoredon
a 5point Likert scale
ranging from 1 = no
Method
discomfortto5=very
muchdiscomfort.Inthe
Item Pool Generation and Initial Item
Frequency scale the
Selection
same items are scored
The original item pool of the Inventory of ona
InterpersonalSituations(IIS)consistedof765items
drawn from the Dutch and English written
inventories on social anxiety and/or assertiveness
available since 1975 (see Van DamBaggen &
Kraaimaat, 1987a) as well as from own clinical
practice. Duplicate itemswereremovedfrom this
item pool, as were all items which did not meet
general criteria for scale construction. The item
criteriawereitsrelevancetotheconceptofsocial
anxiety, unambiguous content, descriptive of
behavior,formulationasanactivebehavior,length
of20wordsorless,devoidofdoublenegatives,and
formulationinsimpleunambiguouslanguage.The
remaining66itemswereclassifiedbyexperienced
behavior therapists according to their content
(situation facet) and appeared to sufficiently
represent the two main domains of positive and
negative assertiveness (e. g., Pitcher & Meikle,
1980). More specific classification showed an
approximately equal representation of the 16 a
prioridistinguished social responses used in our
socialskillstrainingwithpsychiatricpatients(Van
DamBaggen&Kraaimaat,1986).Therandomized
items with instructions were again judged by
clinicalpsychologistsandlinguistsonthebasisof
readability, ambiguity, complexity, and
comprehensibility. Finally, the inventory was
completed by psychiatric patients with low
educationallevelstotestthecomprehensibilityof
theinstructionsandtheitems.
The reaction facet was elaborated in the
construction of two a priori scales: (1) the

5point Likert scale

ranging from 1 = I
never do, to 5 = I
always do. The
sequence of the items
is the same on both
scales.
Item Reduction
Phase
The provisional 66item
versionwasaddedtothe
pretreatmentassessment
of a Social Skills
Training (SST: Van
DamBaggen

&
Kraaimaat, 1986) in a
Dutch psychiatric unit
and completed by
patients who had been
consecutively referred
by clinicians and
psychiatrists. These
patientshadtofulfillthe
criteria formulated for
participationintheSST,
namely,agebetween18
and 65, and social
anxiety and/or social
skill deficits reported
during the behavioral
assessmentinterviewand
on selfreport inven
tories, such as the SAS
(Willemsetal.,1973)for
theemotionalaspectand
theWLAS(Wolpeetal.,
1966)forthebehavioral
aspect. Patients with
various kinds of social
inadequacy, such as
socialanxiety,avoidance
of social situations, and
having deficits and
excesses in social re
sponses, were admitted
to the SST. The nature
and intensity of other
complaints were not
determiningfactorswith
respecttotheadmittance
totheSST.Thisresulted
inatargetsampleof124
socially

anxious
psychiatric patients, 59

27

menand65women,withameanageof32yearsand ttests were used to

the
allofCaucasianethnicity.Thepsychiatricdiagnosesexamine
rangedfromseverelyneurotictoborderlinepsychoticdiscriminative validity
syndromes;patientswithacutepsychosisandorganic of the items either on
disturbances were excluded. In addition, theDiscomfort or on
provisional version was administered to a control Frequency. Using a
sample of 131 normal adult Dutch volunteers, criterion of p < .05, it
employeesofseveraloffices,87menand44women, appeared that all the
withameanageof32.7yearsandallofCaucasian itemsoftheDiscomfort
ethnicity. In both samples the IIS was individually scale discriminated the
administeredbyaresearchassistant.Forthepatients, groups; with the Fre
the IIS was included as part of the pretreatment quency scale, the 13
assessmentoftheSST.Thesamplesdidnotdifferin itemsthatdidnotwere
2
age(ttest:p=.60)oreducationallevel( :p=.12). removed.

Results and Discussion

In order to remove
ambiguous items,
principal components
analyses (varimax
EJPA15(1),1999

rotation and Kaiser

normalization) were
performed on the
remaining 53 items of
both the Discomfort
and Frequency scale
with the data of the
socially

anxious
patients. Based on the
factor analysis of
Frequency, those items
thatloadedhighontwo
or more factors were
removed by using a
criterion of > .40, as
were those that loaded
toolowonafactorby
usinga

Hogrefe&Huber

Publishers

28

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)
to borderline psy
chotic syndromes;
done by computing
patients with acute
criterionof<.50.Next,thesamewasdoneon Cronbachs sandalso
psychosis and organic
thebasisofthefactoranalysisofDiscomfort. correlations comparing
disturbances were
Finally,35definiteitemsremained,whichwere each item with the total
excluded.Fiftypercent
againrandomized.Theinstructionsandformat score of the remaining
of the sample were
of the two scales remained unchanged. Theitems for both the
educated below high
minimum score of each scale is 35 and the Discomfort

and
school. The patients
maximum is 175 (Van DamBaggen &Frequencyscales.
who participated in
Kraaimaat, 1987a, 1990a, 1990b). The
this study were
instructionsand the 35itemsare giveninthe
differentfromthoseof
Method
Appendix.
the itemreduction
Subjects
phase.

Normal
adults (N =
Theinternalconsistency
276),130menand146
wasinvestigatedintwo
women with a mean
samples:
ageof38.6years(SD
Socially anxious
= 13.4; range 16 to
inpatients

and
ThepurposeofStudy2wastoinvestigatethe outpatients (N = 217)
74), employees of an
temporalstabilityoftheIISDiscomfortand from two Dutch
office, participants in
Frequency scales. The stability was psychiatric units, 79
adult educational
investigated with a testretest interval of 6 men and 138 women
servicesorathletes,all
weeksinanonclinicalsample.
Dutch volunteers of
withameanageof32
Caucasian ethnicity,
years(SD=9.2;range
were recruited to
16 to 60) and of
Method
participate in the
Caucasian ethnicity.
study.Thedistribution
These patients were
Subjects
of the educational
participantsofaSocial
Subjectswere53membersofachoir,20men
levels was nearly
Skills Training (SST:
and33women,withameanageof43.1years
symmetrical. The
Van DamBaggen,
(SD=16;range17to73)andallofCaucasian
normalsubjectsofthis
1984; Van Dam
ethnicity.Thesubjectswerevolunteersrecruited
study were different
Baggen & Kraaimaat,
to participate in the study and were different
from those of the
1986), who were
fromthoseoftheitemreductionphase.
previousstudies.
consecutively referred
by

clinicians
Procedure
(psychiatrists and
The IIS was individually administered to the clinical psychologists)
subjectsbyaresearchassistantandcompleted and had to meet the
twiceoveratimeinterval of6weeks.Atthe criteria for the SST.

psychiatric
initial test, the participants were not informed The
abouttheretest.Notreatmentinterventionwas diagnoses ranged
fromseverelyneurotic
administeredbetweentestandretest.

Study 2: Temporal
Stability

4.4) and M =
17.3 (SD =
4.9). Theprod
Results and
uctmoment
Discussion
correlations
The means (standardbetween test
deviations) for test and retestand retest were
for Discomfort were,r=
respectively, M = 19.2 (SD =.84andr=.86

the
5.4)and M =18.9(SD=5.7)for
and for Frequency,Discomfort
respectively, M = 17.0 (SD =and Frequency

scales,

respectively,
ency
which can be
seen

as
The purposeof
relatively high
Study3wasto
reliability
investigate the
overtime.
homogeneity
of

the
Discomfort and
Study 3: Frequency
scales. This
Internal was

Consist

Procedure
The IIS was
individually
administered
to the subjects
by a research
assistant and
was included
as part of the
SST
pretreatment
assessment for

thepatients.

patientsandthe
normal
subjects,
respectively,
Results and
and for the
Discussion
Frequency
The Cronbachs s for thescale .92 and .
Discomfortscalewere.96and.91,respectively
93 for the socially anxious(see Table 1).
EJPA15(1),

These are highBernstein,

for research1994). In
purposes andaddition,itwas
group
shown that
comparisonandalmost all 35
sufficiently
separate items
highforappliedwere related
settings
with the rest
(Nunnally &totalscorewith

1999Hogrefe& HuberPublishers

the socially
anxiouspatients
and also with
the normal
subjectsforthe
Discomfort and
the Frequency
scales. The
item

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)

Table 1. Means, standard deviations and Cronbachs s for the IIS scales in 3 samples.

1. Soc. anxious patients

2. Psychiatric patients
3. Normal subjects

Discomfort
SD
M

217
363
276

101.3 27.0
91.7 28.8
66.8 27.0

.96
.95
.93

ratingsofbothsamples.
Significant
remainder correlations of the Discomfort scaleassociations between
rangefrom.37to.76forthepatientsandfrom.30 groupsofr=.93andr
to .67 for the normal persons, while for the =
Frequencyscaletheyrangefrom.21(item3)to.70 .93, respectively, were
forthepatientsandfrom.30to
revealedforDiscomfort
.57forthenormalpersons.Theseresultssupport and Frequency. This
thattheempiricallyderivedIISitemscontributemeans that, although
totheconstructsmeasuredwiththeDiscomfortthesociallyanxious

andFrequencyscales.

Study 4: Item and Factorial

Invariance
The purpose of Study 4 was (1) to assess the
invariance of the rank order of item means by
empirically investigating the agreement in rank
ordering of twosamples differingwith respect to
socialanxiety,and(2)toassesstheinvarianceof
thefactorialstructurebytestingtheconsistencyof
the a priori clustering of items (situation facet)
acrossthetwosamples(VanDamBaggen,Kraai
maat&Kiers,1991;1992).

Method
Subjects
This study was performed with two samples
representing the extremes of a social anxiety
continuum.ThesamplesofStudy3werealsoused
here:217sociallyanxiouspsychiatricpatientsand
276normaladultpersons.

Results and Discussion

Item Invariance: Rank Ordering of the Items
The agreement of the rank ordering of the items
between the socially anxious patients and the
normalsubjectswithrespecttodiscomfortaswell
as frequency was investigated by computing the
Spearmancorrelationbetweenthegroupmeanitem

Frequency
SD
M
93.7 17.4
96.6 20.0
113.0 16.3

.92
.92
.91

psychiatric patients
and normal subjects
differed in their level
of reported social
anxiety and social
responses (see Table
1), they did agree in
their rank ordering of
the social responses
thatproducedinthem
more or less social
anxiety or of those
items that were
performed more or
lessfrequently.
Factorial Invariance:
Stability of Item
Clustering
The consistency of the
clustering of items of
the two samples was
investigated with the
Simultaneous Com
ponentsAnalysis(SCA:
Millsap & Meredith,
1988; Kiers & Ten
Berge,1989;TenBerge
& Kiers, 1990) for
reported discomfort as
well as response
frequency.

SCA
computes components
asweightedsumscores
ofthevariablesasisthe
case in the Principal
Components Analysis
(PCA),butusesexactly
the same weights for
this computation with
the purpose of exactly
measuring the same
constructs in the
samples.

The
component weights

29

shouldcontributetoanoptimalrepresentationofthecomponents

ex
variablesbythecomponents.
plaining 60.5% and
SCA of the Discomfort scale showed a rather58.2% of the variance
clear structure of five components after oblique forthepatientsandthe

subjects,
rotation, explaining 61.6% and 55.7% of the normal
variance for the socially anxious patients and the respectively. With
normalsubjects,respectively.WithseparatePCAs, separate PCAs, the
the explained variances would have been only explained variances
scarcely higher, that is, 62.1 and 56.1%, re would have been only

higher,
spectively. The five components represent thescarcely
following domains of social behavior: Expressingnamely, 61.2% and
criticism and opinion, Giving compliments, 58.9%, respectively.
Initiating contacts, Positive selfstatements andThe eight components
representthefollowing
Doingandrefusingrequests.
SCAof the Frequency scale showed, also after domains of social
oblique rotation, a rather clear structure of eight behavior: Giving
EJPA15(1),1999

criticism, Expressing
opinion,

Giving
compliments, Initiating
contacts, Positive self
statements,Standupfor
yourself, Doing a
requestandRefusinga
request.
Therefore,itappears
that with similarly
definedcomponentsin
both samples, almost
the same variance can
be explained as with
PCA, while these
simultaneous

Hogrefe&Huber

Publishers

30

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)

Subjects
componentsalsoplayanimportantroleinthe Inthisstudythe
descriptionofthevariablesinbothsamples. followingsampleswere
In order to investigate whether theinvolved:
componentsbehaveinthesamewayinboth the 217 socially
samplesinotherwords,whethertheyalmost anxious psychiatric
have the same loadings we compared the patients ofStudy3as
loadings of each component with Tuckers phi criterionsample,
coefficient. This coefficient measures the the 276 normal adults
agreement between the loadings on two of Study 3 as
components.TheTuckersphicoefficientswere. noncriterion sample
99,.94,.98,.97,and.98,respectively,forthe and
componentsofDiscomfort,and.98,.98,.97,. a general sample of
98,
363 psychiatric patients
.98, .97, .97, and .88, respectively, for the asnon
components of Frequency. This means that
both samples show a similar structure or
clustering for discomfort as well as for fre
quency.
Inspectionofthecontentofthecomponents
revealedthat theclusterssufficientlyrepresent
theclassesofsocialresponsesonthebasisof
whichitemsweregeneratedinthedevelopment
phase.Thissupportsthecontentvalidityofthe
IIS as well as the use of situational facet
approachingeneratingitemsfortheinventory.

criterionsample;152
menand211women
with a mean age of
35.5 years (SD =
10.7;range17to69)
and of Caucasian
ethnicity; about 57%
of the sample were
educatedbelowhigh
school level. This
sample consisted of
patients participating
inastudycomparing
the effectiveness of
inward versus day
hospital treatment.
Only patients who
werenotabletocom
plete

the
questionnaires
because of acute
psychoses were
excluded from the
sample. This sample
did not overlap with
thesampleofsocially
anxious psychiatric
patients.

Study 5: Discriminative
Validity

Procedure

ThepurposeofStudy5wastoinvestigatethe
discriminativeorknowngroupsvalidityofthe
IIS by comparing a sample representing the
criterion social anxiety with samples not
representingthiscriterion.AstheaimoftheIIS
was to measure social anxiety in psychiatric
patientsacrossvariousdiagnosticcategories,the
sampleofselectedsociallyanxiouspsychiatric
patients served as criterion sample. The IIS
scales needed to discriminate the socially
anxiouspsychiatricpatientsamplefromboththe
general sample of psychiatric patients and the
normal subjects. Moreover, the IIS scales
needed to discriminate the general sample of
psychiatricpatientsfromthenormalindividuals
because a relationship between social anxiety
andpsychiatricsyndromeswasfoundinseveral
studies (Bryant, Trower, Yardley, Urbieta, &
Letemendia, 1976; Curran, Miller, Zwick,
Monti,&Stout,1980;Zigler&Glick,1986).

Results and
Discussion

Method

Thepsychiatricpatients
completed the IIS as
part of the pretest
assessment

for
admittance to the in
ward or dayhospital.
The assessment was
conductedbyaresearch
assistant.

Table1givesthemeans
and standard deviations
of the Discomfort and
Frequency scales for the
three samples. The three
groups were compared
withonewayanalysesof
variance, while Student,
Newman,andKeulstests
were used to detect
differences between

pairs.ItappearedthatthethreegroupsdifferedboththeIISscalesshow
significantlyontheDiscomfortscale(F(2,853)agooddiscriminativeor
=131.3; p <.001)andtheFrequencyscale(Fknown groups validity.
(2,853)=88.0; p <.001).Student,Newman,With respect to the
andKeulstests(p <.01)revealedthatthesoseparateitems,theresults
cially anxious patients differed significantlywere similar: All

items
fromthenormalsubjectsonboththeDiscomfortDiscomfort
discriminate

socially
andtheFrequencyscale.Thesociallyanxious
patients also differed significantly from theanxious and normal
psychiatricpatients onDiscomfort,but not onsubjects,whileallexcept
Frequency. In addition, the sample of psychi threeFrequencyitems(3,
atric patients differed significantly from the 4and28)alsodidso.
normalsubjectsonbothIISscales,whichagain
strengthens the construct validity of the IIS.
From these findings it may be concluded that
EJPA15(1),1999

Study 6:
Convergent
and
Discriminant
Validity
The purpose of Study 6
was to investigate the
nomological network of
the IIS scales by
assessingconvergentand
discriminantvalidity.The
relationoftheIISscales

Hogrefe&HuberPublishers

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)
Thefollowing
measureswereusedin
withothersocialanxietymeasureswasinvestigated thisstudy:
inordertoassessconvergentvalidity.Relationships The Symptom Check
werepredictedonthebasisoftheassumptionthat List
(SCL90:
different measurements of the same construct Derogatis, 1977;
would be related. It was predicted that the IIS Dutch version:
Discomfort scale would correlate moderately to Arrindell & Ettema,
verypositivelywiththesocialinadequacyscaleof 1981)wasusedasan
the Symptom Check List90 (SCL90: Derogatis, index

of
1977; Arrindell & Ettema, 1981) and with the psychoneuroticism.
Social Anxiety Scale (SAS: Willems, Tuenderde IntheDutchversion
Haan,&Defares,1973),whiletheIISFrequency of the SCL90, a
scalewouldcorrelatemoderatelytoverynegatively social inadequacy
with these scales. In order to assess deeply the subscalewasde
convergent validity,the relationshipsbetween the
IISscalesandseveralothersubscalesoftheSCL
90 were investigated, namely, hostility, paranoid
ideationandagoraphobia.Itcouldbeassumedthat
social anxiety constitutes hostility and paranoid
ideation,whilefromtheliteratureitisknownthat
agoraphobiaisoftenaccompaniedbysocialanxiety
(e.g.,Barlow,1993).ItwaspredictedthattheIIS
scales would show moderate relationships with
thesemeasures(Cohen,1988).
Discriminant validity was assessed by
investigating the relationships between the IIS
scalesandnonsocial anxietymeasuresalongwith
the demographic variables. Relationships were
predicted on the basis of the assumption that
measurementsofdifferentconstructswouldnotbe
related.ItwaspredictedthattheIISscaleswould
show low relationships with the InternalExternal
Locus of Control Scale (IE scale: Rotter, 1966;
Andriessen,1972).Becauseintheliteraturehardly
anyrelationshipshavebeenreportedbetweensocial
anxiety and demographical variables, it was
predicted that the IIS scales would show no
relationshipswithsex,age,andeducationallevel.

Method
Subjects
Subjects were the 217 socially anxious
psychiatricpatientsofStudy3.
Procedure
InadditiontotheIIS,thefirstconsecutive110
patients completed several other questionnaires
as part of the pretreatment assessment of the
SST. The assessment was conducted by a
researchassistant.
Measures

rived as well as
subscales

for
hostility, paranoid
ideation, and
agoraphobia
(Arrindell

&
Ettema, 1981). The
validity

and
reliability of the
SCL90 has been
examined with
efficacious results
for several Dutch
adult populations,
including
psychiatric ones
(Arrindell &
Ettema,1981).
The Social Anxiety
Scale
(SAS:
Willems,Tuenderde
Haan & Defares,
1973)wasusedasan
index of negative
selfevaluation in
socialsituations.The
validity of the SAS
has

been
demonstrated in
adultandadolescent
populations
(Willems, Tuender
deHaan,&Defares,
1973).
TheInternalExternal
Locus of Control
Scale (IE: Rotter,
1966;Dutchversion:
Andriessen, 1972)
wasusedasanindex
of selfregulation.
ThevalidityoftheI
E scale has been
sufficiently proven
for experimental as
well as practical
purposes
(Andriessen,1972).

Results and
Discussion
Convergent Validity

31

In line with our prediction, the present study convergent validity of

revealed(seeTable2)thattheDiscomfortscalewas theIISscales.
highly positively associated, and the Frequency
scale was highly negatively associated, with bothTable 2. Product-moment
social anxiety measures (SAS and SCL90 social correlations of the IIS scales
inadequacy); this supports the validity of the IIS with demo-graphic variables
scales. The present study revealed that the corre and social and nonsocial
lationswiththeothersubscalesoftheSCL90are anxiety measures.
significant at a moderate level lower than those
withthesocialanxietymeasures(seeTable2).This
meansthattheIISscalesareslightlyrelatedwith
Convergent validity
somatic complaints, agoraphobia, hostility, and SAS1
paranoid ideation, reflecting the overlap between SCL-90 soc. inadequacy1
anxietyandsocialanxietyandalsosupportingthe SCL-90 paranoid ideation1
EJPA15(1),1999

SCL-90 hostility

SCL-90 agoraphobia
Discriminant validity

Sex
2
Age
2
Educational level

I-E scale

* p < .05, ** p < .01

(one-tailed), Spearman
correlations
1

Only the subjects who

endorsed
all
inventories:
N = 109;
2
Whole sample: N =
217

Hogrefe&Huber

Publishers

.36**

.20

.38**

.27**

.15*
.05
.00

.02
.04
.09

.23*

.24*

32

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)

Subjects
Discriminant Validity

Fortyseven inpatients
This study (see Table 2) showed that the and outpatients from a
correlations of the IIS scales with the IEDutch psychiatric
scale were low, which means that the IIS hospital,25menand22
scalesapparentlymeasureotherconceptsthan women,withameanage
theIEscale.Inaddition,itwasrevealedthatof35.1(SD=10.9;range
the Discomfort scale was not significantly20 to 56) and of
relatedwithageandeducationallevel;alow Caucasianethnicity,were
butsignificantcorrelationwasfoundwiththe recruited to participate
Discomfort scale with respect to sex. Novoluntarily in an
significantcorrelationswerefoundwithsex,experimental study on
age,andeducationallevelwiththeFrequencyaspectsofsocialanxiety.

psychiatric
scale, which means that this scale isThe
diagnoses

ranged
from
independent of these characteristics. These
severelyneurotictobor
findings support the discriminant validity of

theIISscales.

Study 7: Predictive
Validity
ThepurposeofStudy7wastoassessthedegree
towhichscalescoresoftheIIS couldpredict
social anxiety related to overt behavior in a
naturalistic social situation. The subjects were
exposed to a social situation in a naturalistic
roleplay procedure. The question of the
predictive validity is important because the
predictive value of selfreport inventories for
overtbehaviorwasoftenfoundtoberatherlow
(Mischel,1968,1973,1990).Inthefirststageof
thedevelopmentoftheIISwetriedtosolvethis
problembyformulatingtheitemsoftheIISas
social responses to specific situations.
Therefore,weexpectedtheIISscalestohavea
relativelyhighpredictivevalue.
Amoderatetohighcorrelationbetweenthe
IISscalesandovertbehaviorinsocialsituations
was required for the predictive validity. More
specifically,itwasexpectedthattheDiscomfort
scale would show negative correlations with
overtbehavior(withtheexceptionofresponse
latency),whiletheFrequencyscalewouldshow
positivecorrelationswiththebehavioralaspects
(withtheexceptionofresponselatency).

Method
For more extensive information about the
selection of the subjects, the procedure and
the measures, see Van DamBaggen and
Kraaimaat(1987b).

derline

psychotic
syndromes; patients
with acute psychoses
and

organic
disturbances were
excluded from the
experiment. The patient
sample of this study is
different from those
usedinthepriorstudies.
Procedure
TheIISwasadministered
to the subjects by a
research assistant two
weeks before the
experiment. During the
experiment

the
participantswereexposed
toanaturalisticroleplay
social situation with a
confederate. The sub
jects were instructed to
initiate a conversation
withanunfamiliarperson
inawaitingroomandthe
confederate

was
instructedtoreinforcethe
subjects efforts without
taking any initiatives.
Theroleplayingcameto
an end in two minutes.
During the situation the
subjectsovertbehavioral
reactionstothesituation
wererecorded.

Measures
Overt behavior was
continuously recorded
duringtheexperimental
sessions bymeans of a
video recorder. In
dependent judges,
unfamiliar with the
experimental design,
scored and rated the
following types of
behaviorfromthevideo
andaudiotapes:
The duration of
speech: the total time
of speech in minutes
during the first five

responses;
The content of the
The response latency: the mean time (in verbal response: this
seconds) between the end of the was rated with the
confederatesresponseandthebeginningof helpofthewrittentext
theparticipantsresponse;
and a 7point Likert
Thenumberof clauses: thetotalnumberof scale with respect to
kind and variation of
verbalresponsesperminute;
Thedurationofgaze: thetotaltimeofgaze the responses in

the
during the first five reactions of the initiating
conversation.
participant;
Theadjustmentofgaze:theadjustmentofthe
variationingazetotheinteraction(7pointPrevious research
revealed sufficient
Likertscale);
Volumeofvocalization: thetuningofspeechvalidity for these

in
volume to the interaction (7point Likertmeasures
discriminating high
scale);
Intonation: the adjustment of variation infrom low socially
anxious patients (Van
intonation(7pointLikertscale)and
EJPA15(1),1999

DamBaggen

&
Kraaimaat,1987b).
As a check on
reliability, 25% of the
video tapes were
randomly rescored and
rerated by independent
judges unfamiliar with
thedesign.Theinterrater
reliability

scores
(productmoment
correlations)werespeech
duration r = .99,
responselatency r =.95,
number of verbal
responses r = .98,
durationofgazer=0.99,
adjustment

Hogrefe&HuberPublishers

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)
predictive validity of
the IIS Discomfort
ofgaze r =.78,volumeofvocalization r =.79,and Frequency scales
andintonation r =.66.Withanintervaloftwo is moderate to high
weeksthecontentwasratedtwicebyaclinical with respect to the
psychologist/behavior therapist who was very group of overt social
experienced in the treatment of social anxiety. behaviors.
The intrarater reliability (productmoment cor
relation) was r = .94 (Van DamBaggen &
Kraaimaat,1990).

Results and Discussion

First, a check was done on the social anxiety
levelofthepatientsample.Itwasfoundthatthe
meanscoresofDiscomfort(M=93.6,SD=28.3)
and Frequency (M = 97.3, SD = 19.5)
corresponded more or less with the mean of a
reference group of heterogeneously diagnosed
psychiatric patients (Van DamBaggen &
Kraaimaat,1987a)(cf.Table1).Table3presents
the correlations of the predictive variables and
theIISscales.Onetailedtestswereusedtotest
thesepredictions;theBonferronicorrectionwas
usedtocontrolfortestwiseerrorintheanalyses.
Table 3. PM-correlations and multiple correlations
of overt behav-ior and IIS scales.

Overt behavior
Duration of speech
Response latency
Number of clauses
Duration of gaze
Adjustment of gaze
Volume of vocalization
Intonation
Content

Discomfort
2
R
r

Frequency
2
R
r

.19
.23
.30
.51*
.42*
.40*
.18
.32

.31
.43*
.42*
.45*
.46*
.44*
.37*
.41*

.62
.58
Set of overt behaviors
*p < .006 (after Bonferroni correction; one-tailed),

p < .01 (one-tailed)

EightpredictionsweremadeforeachIISscale.
Three out of eight correlations were found to be
significantfortheDiscomfortscale,whilesevenout
ofeightweresignificantfortheFrequencyscale.
This is also reflected in the multiple regression
coefficients,whichwerehighandsignificantforthe
set of overt behaviors. Thus, in general, the
Discomfort as well as the Frequency scale were
foundtopredictovertbehaviorinasocialsituation.

Fromtheseresults,itcanbeconcludedthatthe

Study 8:
Sensitivity
to Change
ThepurposeofStudy8
was to determine the
degreetowhichtheIIS
was sensitive to
treatmentrelated
changes in social
anxiety. The IIS was
administeredtothesub
jects of a validated
Social Skills Training
(SST) at pretreatment
and posttreatment. In
previous research, the
effectiveness of the
SST had already been
established with other
outcomemeasuresthan
the IIS (Van Dam
Baggen, 1984; Van
DamBaggen

&
Kraaimaat,1986).This
SSThastheformatofa
broad

spectrum
treatment program,
directed at emotional,
cognitive,

and
behavioral aspects of
social behavior, and
aimed at the reduction
of social anxiety and
the enhancement of
social skills as well as
the enhancement of
selfregulative
behaviors.

Method
Subjects
This study on the
sensitivity of the IIS
was performed with
three nonrandomized
samples of socially
anxious psychiatric
patients:
Socially anxious
psychiatric patients

33

(N = 136) who completed the SST in an

outpatientsetting,85womenand51menwitha The exclusion criteria
meanageof30.7(SD=9.5)andofCaucasian usedwiththesepatient
ethnicity;
samples were similar
Sociallyanxiouspsychiatricpatients (N =102)to those in the
whocompletedtheSSTinaninpatientsetting,aforementioned
67womenand35menwithameanageof30.4studies. The patient
(SD=9.4)andofCaucasianethnicity;
samples of this study
Socially anxious psychiatric patients (N = 28)were different from
whofulfilledthecriteriaforSST.Onlygeneralthe patient samples
psychiatric treatments (e. g., medication,used in the prior
sociotherapy,occupationaltherapy)weregiven validity studies. Fur
tothesepatients,butnottheSST.Thissampleof thermore,itshouldbe
patientsformedtheTreatmentControlConditionnoted that it was
(TCC) and consisted of 20 women and 8 men impossible to study
with a mean age of 36.6 (SD = 9.3) and of sensitivity to change
in untreated patient
Caucasianethnicity.
EJPA15(1),1999

samplesbecauseofthe
ethical constraints in
withholdingtreatment
from psychiatric
patients.
Procedure
The IIS was
completed as part of
the pretreatment and
posttreatment
assessmentofanSST
in the case of the
treatedsamples,while
pretests and posttests
werecom

Hogrefe&Huber

Publishers

34

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)

This paper has focused

on the development and
pletedwithatimeintervallastingaslongas
the SST in the control condition. The validation of a self
assessments were conducted by a research reportinstrumentforthe
assessment of social
assistant.
anxiety,theInventoryof
Interpersonal Situations
(IIS). The questionnaire
Results and Discussion
consists of 35 items
InTable4,meansandstandarddeviationsareformulated as responses
given of pretests and posttests for theto specific social
DiscomfortandFrequencyscales,aswellas tsituations.Itisbasedon
values of the comparison of pretests and aninteractiveconceptof
posttests and their effect sizes (Cohen, 1988).social anxiety and
PretestsandposttestsofthetwoSSTconditions provides scales for
werefoundtobesignificantlydifferentforboth discomfort (emotional
scalesoftheIIS.Thepretestsandposttestsof aspect)andforfrequen
thecontrolconditionalsodifferedsignificantly.
TheseresultsindicatethatthescalesoftheIIS
are sufficiently sensitive to detect changes
causedbyaspecifictherapyaimedatbothsocial
anxietyanddeficitsinsocialskillsaswellasby
generalpsychiatrictherapies.Theeffectsizesof
the SSTs are rather large, which strongly
suggeststhattheIISaccuratelydetectschanges
in the domain of social anxiety. As could be
expected, the effect sizes of the control
conditionaresmallerbecausethesocialanxiety
level inthisconditionisonlyindirectlyinflu
encedbythegeneral psychiatrictreatments.It
shouldbenotedthattheseeffectsareaboutthe
same as those of both common factors and
placebo effects (Lambert, Shapiro, & Bergin,
1986).However,itcouldalsobehypothesized
that the rather small effect size on the
Discomfort scale should be attributed to a
bottomeffectcausedbytheratherlowlevelof
thepretest.
Table 4. Means and standard deviations of
pretests and posttests, t-values and effect sizes
of the three samples.

Pretest
Scale M
SD

Group N
SST-a1 136
SST-b1 102
TCC2

28

Posttest
M
SD

Disc. 105.2 28.6 78.2 24.2 11.7** 1.02

Freq. 92.2 17.5 109.9 19.8 10.6**
Disc. 103.1 25.2 79.0 21.8 9.5**
Freq. 95.0 13.2 110.1 18.2 8.3**
Disc. 92.3 28.7 82.5 24.8 2.2*
Freq. 91.2 19.4 100.1 18.4 3.9**

*p < .05, **p < .01 (two-tailed)

1

SST = Social Skills Training; 2TCC = Treatment Control

Condition

General Discussion

cy (behavioral aspect).
Because the assessment
of the psychometric
characteristics of an
instrument is an on
going process requiring
multiple efforts over
time, a set of 8 studies
waspresentedtotestthe
validity and reliability
oftheIIS.
Social anxiety as
measured by the IIS
scales Discomfort and
Frequency showed good
internal consistency
across multiple clinical
and nonclinical groups.
In addition, it was
revealedthattheinternal
structure was rather
invariant across socially
anxious and nonsocially
anxious

groups.
Although these groups
differed in their level of
reported social anxiety,
theyagreedintheirrank
ordering of the social
responsesthatgavethem
relatively more or less
anxietyaswellasintheir
clustering of social
responses.
The results of the
discriminative validity
study indicate that the
IIS scales significantly
distinguish between
psychiatric patients who
werereferredtoasocial
anxiety intervention
group and normal
persons.Itshouldbenot
ed that the Discomfort
scale had a particularly
high discriminative
value. The Frequency
scale tended to dis
criminate the socially
anxiouspatientsfromthe
general sample of
psychiatricpatients,butit
did discriminate both
patient groups from the

normal persons. One explanation for thishigh predictive value.

finding might be that reported social anxiety From predictive as well
functionedmoreasacriterionforassignmentto asdiscriminativevalidity
treatmentthandidreportedfrequency.Anotherresults it can be
explanation could be found in the differentconcluded that the
mechanisms which were at the basis of lowDiscomfort

and
frequencyscoressuchasbehavioralinhibitionFrequency scales indeed
duetosocialanxietyandlackofsocialskills. tapcrucialdimensionsof
socialanxiety.
ThepurposeofthedesignoftheIISscales Social anxiety as
was to develop an instrument with high measured by the IIS
predictive qualities for specific behaviors inscales proved to be
specificsituations.Thefindingthatbothscales relatively independent
predict overt behaviors in a conversational of sex, age, and
situation demonstrated that early shortcomingseducationallevel,which
ofselfreportinventoriescanbeadequatelymet.is congruent with the
Inouropinion,theformulationoftheitemsinaliterature. The signif
concrete and situationspecific manner mayicant, but very small,
havesubstantiallycontributedtothisrelatively
EJPA15(1),1999

correlation of the
Discomfort scale with
sex could be due to
differences in specific
domains of social
anxiety between men
andwomen,asissome
times reported in the
literature (e. g., Wilson
& Gallois, 1993); this
needs

further
explorationwithrespect
totheIIS.
FortheuseoftheIIS
in clinical practice the
individualsscorescould
be compared with those
ofreference

Hogrefe&HuberPublishers

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)

References

groups, as there are socially anxious psychiatric

patients, heterogeneous psychiatric patients, and
normal persons (Van DamBaggen & Kraaimaat, Alden, L., & Cappe, R.

Non
1990b).Thiscomparisoncouldserveasthebasis (1981).
assertiveness:

skills
fortheselectionfortreatment.Itisofinterestto
deficitorselectiveself
note that the following productmoment
evaluation? Behavior
correlations were obtained for the relationship Therapy,12,107114.
betweentheDiscomfortandtheFrequencyscale:r AmericanPsychiatric
=.62forthesociallyanxiouspsychiatricpatients, Association(1980,1987,
r=.49fortheheterogeneouspsychiatricpatients, 1994).Diagnostic
and r =.43forthenormal persons.However,a
treatment preference for social anxiety cannot be
inferredfromthescoresoftheIISscales;ratheritis
dependent on the mechanisms underlying the
patientsexperienceddiscomfortandperformance
ofsocialbehavior(VanDamBaggen,Kraaimaat,
&Crouzen,1993).
Thestudyontheitemandfactorialinvarianceof
theIISrevealedrelativestablerankorderingaswell
as clustering of social responses on Discomfort
andFrequencyacrosstwosamples.Thismeansthat
therelativeanxietyelicitingcapacityofresponses
insocialsituationsisratherstable.Inaddition,the
clustersofsocialresponsesinferredforDiscomfort
andFrequencyhighlyagreeacrossthetwosamples.
Itwasalsorevealedthatthecomponentsaresimilar
tothefactorial dimensionsfoundinotherstudies
with different samples (cf. Van DamBaggen,
Kraaimaat,&Kiers,1991,1992).
The results of this set of studies support the
validity and reliability of the IIS Discomfort and
Frequency scales. The IIS scales should be very
useful for differentiating socially anxious
psychiatric patients, heterogeneous psychiatric
patients,andnormalsubjects.Furthermore,theIIS
scaleshaveahighpredictivevalueforovertaspects
of social behavior, which is important for use in
clinicalpractice.Inaddition,theIISscalescanbe
usedclinicallytohelpdelineatespecificaspectsof
socialanxietyinaspecificindividual.Thiscanbe
donebyperusingthesituationalandresponsefacets
assessedbythescales.Finally,theIISscalesshould
prove to be a useful instrument to measure
treatment outcome. In future research the
significance of the IIS should be investigated for
crossculturalstudiesinsocialanxiety.Inaddition
totheDutchversion,theIISiscurrentlyavailable
in both British and American English, French,
German,andTurkishtranslations.CopiesoftheIIS
for research purposes are available from the first
author.

andStatisticalManual
ofMentalDisorders
DSM(3rd,3rdrevised,
and4thedition).
WashingtonDC:
Author.
Andriessen, J. H. T.
(1972). Interne of
externe beheersing
[Internal

versus
external

control].
Nederlands Tijdschrift
voor de Psychologie,
27,173198.
Angleiter,A.,John,O.P.,&
Lhr, F. J. (1986). Its
what you ask and how
youaskit:Anitemmetric
analysis of personality
questionnaires. In A.
Angleiter&J.S.Wiggins
(Eds.),
Personality
assessment

via
questionnaires. Berlin:
SpringerVerlag.

Arrindell, W. A., &

Ettema, H. (1981).
Dimensionele structuur,
betrouwbaarheid en
validiteit van de
Nederlandse bewerking
van de Symptom
Checklist (SCL90);
gegevens gebaseerd op
een fobische en een
normale populatie
[Dimensionalstructure,
reliability and validity
oftheDutchversionof
theSCL90:Databased
on a phobic and a
normal population].
Nederlands Tijdschrift
voor de Psychologie,
36,77108.
Asendorpf, J. B. (1988).
Individual response
profiles in the behav
ioral assessment of
personality. European
JournalofPersonality,
2,155167.
Barlow, D. H. (1993).
Clinical handbook of
psychological disor
ders. New York:
Guilford.
Beidel, D. C., Turner, S.
M., & Dancu, C. V.
(1985). Physiological,
cognitiveandbehavioral
aspects of social
anxiety.
Behaviour

35

Stout, R. L. (1980). The

ResearchandTherapy,23,109117.
Bruch,M.A.(1981).Ataskanalysisofassertivebehaviorsocially inadequate
revisited:Replicationandextension. BehaviorTherapy, patient: incidence rate,
demographical and
12,217230.
Bruch,M.A.,Gorsky,J.M.,Collins,T.M.,&Berger,P.A. clinical features, hospital

posthospital
(1989). Shyness and sociability reexamined: aand
multicomponent analysis. Journal of Personality and functioning. Journal of
Consulting and Clinical
SocialPsychology,57,904915.
Bryant, B., Trower, P., Yardley, K., Urbieta, K., &Psychology,48,375382.
Derogatis, L. R. (1977).
Letemendia,F.J.(1976).Asurveyofsocialinadequacy
among psychiatric patients. Psychological Medicine, 6,SCL90: Administration,
scoring and procedure.
101112.
Clark,J.V.,&Arkowitz,H.(1975).Socialanxietyandself Manual I for the

version.
evaluationofinterpersonalperformance. Psychological r(evised)
Baltimore:John Hopkins
Reports,36,211221.
Cloitre,M.,Heimberg,R.G.,Holt,C.S.,&Liebowitz,M.R. University School of
Clinical
(1992).Reactiontimetothreatstimuliinpanicdisorderand Medicine,
Psychometrics
Research
social phobia. Behaviour Research and Therapy, 30, 609
Unit.
617.
Gambrill, E. D., & Richey,
Cohen, J. (1988). Statistical power analysis for the
C. A. (1975). An
behavioralsciences(2nded).Hillsdale,NY:Erlbaum.
assertion inventory for
Curran,J.P.,Miller,I.W.,Zwick,W.R.,Monti,P.M.,&
EJPA15(1),1999

use in assessment and

research.
Behavior
Therapy,6,550561.
Glass, C. R., Merluzzi, T.
V., Biever, J. L., &
Larsen, K. H. (1981).
Cognitiveassessmentof
social

anxiety:
development and val
idation of a self
statementquestionnaire.
Cognitive Therapy and
Research,32,255267.
Heimberg, R. G., Chiauzzi,
C. J., Becker, R. E., &
MadrazoPeterson, R.
(1983).

Cognitive
mediation of assertive
behavior: An analysis of
theselfstatementpattern
ofcollegestudents,psy

Hogrefe&Huber

Publishers

36

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)

Nunnally, J. C., &

Bernstein, I. H. (1994).
Psychometrictheory(3rd
chiatricpatientsandadults.CognitiveTherapy
ed., pp. 264265). New
andResearch,7,455464.
York:McGrawHill.
Hersen,M.,Bellack,A.S.,Turner,S.M.,Williams, Pitcher,S.W.,&Meikle,S.
M. T., Harper, K., & Watts, J. G. (1979). (1980). The topography
Psychometric properties of the WolpeLazarus of assertive behavior in
Assertiveness Scale. Behaviour Research and positive and negative
Therapy,17,6369.
situations.
Behavior
Hope, D. A., Rapee, R. M., Heimberg, R. G., & Therapy,11,532547.
Dombeck,M.J.(1990).Representationsoftheselfin Rathus,S.A.(1973).A30
item schedule for
social phobia: Vulnerability to threat. Cognitive
assessing assertive
TherapyandResearch,14,177189.
Behavior
Kiers, H. A. L., & Ten Berge, J. M. F. (1989). behavior.
Therapy,8,393397.
Alternatingleastsquaresalgorithmsforsimultaneous
components analysis with equal component weight Richardson,F.C.,&Tasto,
D. L. (1976).
matricesintwoormorepopulations.
Development and factor
Psychometrika,54,467473.
analysis of a social
Lambert,M.J.,Shapiro,D.A.,&Bergin,A.E.(1986).
anxiety

inventory.
Theeffectivenessofpsychotherapy.InS.L.Garfield
Behavior Therapy, 7,
&A.E.Bergin(Eds.), Handbookofpsychotherapy
453462.
andbehaviorchange
Romano,J.M.,&Bellack,
(pp.157211).NewYork:Wiley.
A.S.(1980).Social
Lang, P. J. (1971). The application of
validationofa
psychophysiological methods to the study of
psychotherapy and behavior modification. In A. E.
Bergin&S.L.Garfield(Eds.),Handbookofpsycho
therapy and behavior change (pp. 75125). New
York:Wiley.

LaVome Robinson, W., & Calhoun, K. S. (1984).

Assertiveness and cognitive processing in
interpersonal situations. Journal of Behavioral
Assessment,6,5970.
Leary, M. R. (1983). A brief version of the Fear of
Negative Evaluation Scale. Personality and Social
PsychologyBulletin,9,

371376.
McFall,M.E.,Winnett,R.L.,Bordewick,M.C.,&
Bornstein,P.H.(1982).Nonverbalcomponentsin
the communication of assertiveness. Behavior
Modification,6,121140.
Millsap,R.E.,&Meredith,W.(1988).Components
analysis in crosssectional and longitudinal data.
Psychometrika,53,123134.
Mischel, W. (1968). Personality and assessment.
NewYork:Wiley.
Mischel,W.(1973).Towardacognitivesociallearning
conceptualization of personality. Psychological
Review,80,252283.

Mischel,W.(1990).Personalitydispositionsrevisited
and revised: A view after three decades. In L.
Pervin(Ed.),Handbookofpersonality.Theoryand
research(pp.111134).NewYork:Guilford.
Monti,P.M.,Boice,R.,Fingeret,A.L.,Zwick,W.R.,
Kolko, D., Munroe, S., & Grunbergen, A. (1984).
Midilevel measurement of social anxiety in
psychiatric and nonpsychiatric samples. Behaviour
ResearchandTherapy,22,651660.
Nelson,R.O.,Hayes,S.C.,Felton,J.L.,&Jarrett,R.B.
(1985).Acomparisonofdataproducedbydifferent
behavioralassessmenttechniqueswithimplications
for models of social skills inadequacy. Behaviour
ResearchandTherapy,23,111.

componentmodelof
assertivebehavior.
JournalofConsulting
andClinicalPsychology,
48,478490.
Rotter, J. B. (1966).
Generalizedexpectancies
for internal versus
external control of
reinforcement.
Psychological
Monographs, 80 (Whole
No.609).
Schene, A. H. (1992).
Psychiatric partial and
fulltimehospitalisation.
Acomparativestudy(pp.
104).Utrecht:University
ofUtrecht.
Schwartz, R. M., &
Gottmann, J. M. (1976).
Toward a task analysis
of assertive behavior.
Journal of Consulting
andClinicalPsychology,
44,910920.
St. Lawrence, J. S. (1982).
Validation of a
component model of
social skill with
outpatientadults.Journal
of Behavioral As
sessment,4,1526.
Stouthard, M. E. A.,
Hoogstraten, J., &
Mellenbergh, G. J.
(1995). A study on the
convergent

and
discriminant validity of
the Dental Anxiety
Inventory.
Behaviour
Research and Therapy,
33,589595.
TenBerge,J.M.F.,&Kiers,
H.A.L.(1990).Simultane
componentenanalyse
voortweeofmeergroepen
personen [Simultaneous
components analysis for
two or more groups of
persons].

Nederlands Tijdschrift
voor de Psychologie, 45,
221226.Trower,P.(1980).
Situational analysis of the
componentsand
processesofbehaviorof
sociallyskilledand
unskilledpatients.

Journal of Consulting
and Clinical Psychology,
48, 327339. Turner,S.M.,
Beidel,D.C.,Dancu,C.V.,

Handleiding bij de
&Stanley,M.A.
InventarisatielijstOmgaan
(1989).Anempiricallyderivedinventoryto
metAnderen,deIOA
measuresocialfearsandanxiety:TheSocial
PhobiaandAnxietyInventory.
[ManualoftheInventory
ofInterpersonal
PsychologicalAssessment:JournalofConsulting
Situations:IIS].Lisse:
andClinicalPsychology,1,3540.
Turner,S.M.,Beidel,D.C.,&Larkin,K.T.(1986). Swets&Zeitlinger.
Situationaldeterminantsofsocialanxietyinclinic Van DamBaggen, R., &
and nonclinic samples: Physiological and Kraaimaat, F. (1987b).
cognitive correlates. Journal of Consulting and Aspectsofsocialanxiety
andsituationalspecificity
ClinicalPsychology,54,523527.
Van DamBaggen, R. (1984). Klinische inpsychiatricpatients.In
sociaalvaardigheidstherapie: Opzet en evaluatie J. P. Dauwalder, M.
[Clinical social skills training: Program and Perrez,&V.Hobi(Eds.),
Controversial issues in
evaluation].Gedragstherapie,17,209223.
behavior modification.
Van DamBaggen, R., & Kraaimaat, F. (1986). A
annual series of
group social skills training program with
european research in
psychiatric patients: Outcome, dropout rate and
behavior therapy (Vol.
prediction. BehaviourResearchandTherapy, 24,
2). Amsterdam/Lisse:
161169.
Swets&Zeitlinger.
VanDamBaggen,C.M.J.,&Kraaimaat,F.W.(1987a).

EJPA15(1),1999

Van DamBaggen, R., &

Kraaimaat, F. (1989).
Socialanxiety:Areview.
InP.M.G.Emmelkamp,
W.T.A.M.Everaerd,F.
W. Kraaimaat, & M. J.
M.vanSon(Eds.),Fresh
perspectives on anxiety
disorders.(AnnualSeries
ofEuropeanResearchin
Behavior Therapy,
Volume

4).
Amsterdam/Lisse: Swets
&Zeitlinger.
VanDamBaggen,C.M.J.,&
Kraaimaat,F.W.(1990a).
Zelfbeoordeling van
sociale angst als predictor
van observeerbaar gedrag
in sociale situaties [Self
reportofsocialanxietyas
pre

Hogrefe&HuberPublishers

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)

Willems, L. F. M.,
dictorofovertbehaviorinsocialsituations].Nederlands
Tuenderde Haan, H.
TijdschriftvoordePsychologie,45,180183.
A., & Defares, P. B.
VanDamBaggen,C.M.J.,&Kraaimaat,F.W.(1990b). (1973).Eenschaalom
InventarisatielijstOmgaanmetAnderen,IOA.Herziene socialeangsttemeten
Handleiding deel 2 [Inventory of Interpersonal [A scale to measure

anxiety].
Situations.IIS.Revised Manual,vol.2].Lisse:Swets& social
NederlandsTijdschrift
Zeitlinger.
VanDamBaggen,R.,Kraaimaat,F.,&Crouzen,M.(1993). voorPsychologie, 28,
De effectiviteit van sociaalvaardigheidstherapie bij drie 415422.
subtypen sociaal angstige psychiatrische patinten [TheWilson,K.,&Gallois,C.
effectivenessofsocialskillstraininginthreesubtypesof (1993).Assertionandits
socially anxious psychiatric patients]. Gedragstherapie, socialcontext.
Oxford:Pergamon.
26,2132.
VanDamBaggen,R.,Kraaimaat,F.,&Kiers,H.(1991).DeWolpe, J., & Lazarus, A.
Inventarisatielijst Omgaan met Anderen (IOA): Een A. (1966). Behavior
vervolgonderzoek naar de validiteit van de subschalen therapy techniques: A
[TheInventoryofInterpersonalSituations(IIS):Afollow guide to the treatment
up study on the validity of the subscales]. of neuroses. Oxford:
Gedragstherapie,24,8998.
Pergamon.
VanDamBaggen,C.M.J.,KraaimaatF.W.,&Kiers,H.A. Zigler, E., & Glick, M.
L.(1992).Eenexploratiefonderzoeknaardestructuurvan (1986).
A
socialeangstbijsociaalangstigepsychiatrischepatienten developmental
[Anexploratorystudyofthestructureofsocialanxietyin approach to adult
socially anxious psychiatric patients]. Nederlands psychopathology (pp.
4367). New York:
TijdschriftvoordePsychologie,47,4145.
Watson, D., & Friend, R. (1969). Measurement of social Wiley.
evaluative anxiety. Journal of Consulting and Clinical
Psychology,33,448467.

Appendix
Instructions for For
Part 1: Discomfort example:
ThisinventoryconsistsofIfyoufeel
anumberofinterpersonala
fair
situations.Pleaseindicateamount of
the degree of discomfortdiscomfort
you would experience inwhen you
each of these situations. join a
Usethefollowinganswerconversati
on of a
key:
small
1. none
group of
people,
2. alittle
thencircle
3. afairamount
number 3
4. much
asfollows:
5. verymuch
situatio
Please complete thentothe
next.
following
questionnaire. TakeThere
yourtimewhenyouare no
work from oneright or

:
of Interpersonal
Inventory Situations (IIS)

1.Joining Instruct
a
ions for
conversatio
Part 2:
nofa
Freque
small
groupof ncy of
people
Occurre
1

2
4
5

RienvanDamBaggen
DepartmentofClinical
Psychology
Universityof
Amsterdam
Roetersstraat15
NL1018WB
Amsterdam
TheNetherlands
Tel.:+31205256817
Fax:+31206391369
Email:
dambaggen@psy.uva.nl
.

nce

following
answers:

1. never
2. seldom
3. sometim
es
4. often
5. always

Inthispartyouwill
find the same 35For
interpersonal situexample:
ationsasdescribed
inPart1.ThistimeIfyou
youaretoindicateneverjoin
how often youa
behave

asconversati
described in theonofa
situations. Use the
wrong thatmatters. One by
sonal there
answers
one
situationare no
; it is
complet
s,takingright or
rather
ethelist
your
wrong
your
of
time. answers
opinion
interper
Again, ;itonly

small
groupof
people,
thencircle
number1
asfollows:
1.Joining
a
conversatio
nofa
small
groupof
people
2
3
4
5

matters
what
you
think.
Take
your

37

time to completePart2.
EJPA15

(1),

1999Hogrefe&

Huber

Publishers

38

R.vanDamBaggenandF.Kraaimaat:TheInventoryofInterpersonalSituations(IIS)

29. Initiating
18. Returning

a
defective

item

(for
Items of the IIS
example, in a store
1. Joiningaconversationofasmallgroupof orrestaurant)
19. Askingforafurther
people
2. Telling a friend that he/she is doing explanation about
something you did
somethingthatbothersyou
notunderstand
3. Resistingpressuretoacceptanoffer(for
20.
Expressing your
example,atthedoor,inthestreet)
opinion in a
4. Accepting a compliment for something conversation with a
youdid
group of unfamiliar
5. Asking a friend to help you with people
something
21. Telling someone
6. Requesting the return of something you thathe/sheoffended
havelenttosomeone
you
7. Turningdownarequesttolendsomeone 22. Refusing a request
money
from a person you
8. Refusing a request from an authority like
figure(e.g.,employer,superior,teacher) 23. Expressing your
9. Tellingsomeonethatyouarepleasedwith appreciation for a
whathe/shedidforyou
present
10. Askingsomeonetostopbotheringyouin 24. Telling someone
apublicplace(theater,subway)
that he/she is good
11. Maintaining eye contact during a looking
conversation
25. Discussing why
12. Asking for information (at a window or someone seems to
booth)
avoidyou
13. Initiatingaconversationwithanattractive 26. Telling someone
male/female
that you likeit that
14. Expressing an opinion that differs from he or she appre
ciatesyou
that of the person with whom you are
talking
27. Agreeing with a
compliment about
15. Initiatingaconversationwithastranger
16. Expressing an opinion that differs from yourlooks
thatofthosearoundyou
28. Telling someone
17. Complimenting someone for a job well thatyouarepleased
with something you
done
did

a
conversation with a
stranger
30. Expressing your
opinionoflife
31. Telling someone
you no longer want
toseehim/her
32. Insisting that
someonecontributes
his/hershare
33. Telling someone
that the way he/she
is talking disturbs
you
34. Expressing your
opinion to an
authority figure (e.
g.,

employer,
superior,teacher)
35. Asking a friend to
gooutwithyou
Pleasecheckifyou
markedallsituations

EJPA15(1),
1999Hogrefe&
HuberPublishers