A Hybri d Model of Soci al Phobi a: An Anal ysi s of Soci al Anxi ety and Rel ated

Symptoms of Anxi ety

Daniel F. Gros,
1,2
Leonard J. Simms,
3
and Martin M. Antony
4
1
Ralph H. Johnson Veterans Administration Medical Center
2
Medical University of South Carolina
3
University at Buffalo, The State University of New York
4
Ryerson University
The categorical underpinnings of the current diagnostic nomenclature have been the subject of
repeated criticism. Recently, researchers have proposed several alternatives to the current system,
including hybrid models of combined diagnostic categories and symptom dimensions. In the present
study, we investigated the symptoms associated with a hybrid model of social phobia. The study
included (a) the development of the initial set of symptom dimensions, generation of an item pool, and
review by an expert panel, and (b) data collection and component analysis of the item pool to determine
the structure of the symptoms. Six separate theoretically relevant symptom dimensions were
identied. Implications of these ndings on the development of a new hybrid model of social phobia
were discussed. & 2010 Wiley Periodicals, Inc. J Clin Psychol 67:293307, 2011.
Keywords: social anxiety; social phobia; component analysis; hybrid model
The Diagnostic and Statistical Manual of Mental Disorders (DSM; American Psychiatric
Association [APA], 2000) has been criticized for organizing psychopathology into discrete
categories, despite the fact that dimensional approaches are more often supported in the extant
psychopathology literature (e.g., Krueger, Watson, & Barlow, 2005; Widiger & Samuel, 2005).
The limitations of the DSM are particularly problematic in the anxiety and mood disorders
because of the high rates of comorbidity and overlapping symptomatology among these
conditions (Brown, Campbell, Lehman, Grisham, & Mancill, 2001). In addition, the polythetic
nature of DSM criteria permits extensive within-category heterogeneity for these disorders
(Krueger et al., 2005), leading to a proliferation of separate conditions within common,
underlying dimensions of psychopathology. This expansion of disorders (e.g., 4 anxiety
disorders in DSM-II to 12 anxiety disorders in DSM-IV) may have contributed to additional
confusion within the diagnostic system as these new diagnoses have been added to account for
gaps between disorders, rather than reecting new discoveries. Together, these issues contribute
to a growing literature on concerns regarding the development of the next revision of the DSM.
The combination of diagnostic concerns and the overlapping nature of current disorder
classes have led some researchers to call for major changes in the DSM, including a
rearrangement of the disorders (e.g., Watson, 2005) and the addition of dimensional models of
symptomatology (Mineka, Watson, & Clark, 1998). Combining these two proposed changes,
Brown and Barlow (2005) argued for a hybrid approach to diagnosis, incorporating dimensional
ratings of diagnostic categories and symptom criteria into the current or revised diagnostic
categories. These dimensions may involve both diagnostic severity ratings within each diagnostic
category and higher order constructs that may account for general risk factors and comorbidity
across disorders. Although a categorical-dimensional hybrid approach still may result in

This article was reviewed and accepted under the editorship of Beverly E. Thorn.
This research was supported in part by a grant from the Mark Diamond Research Fund of the State
University of New York at Buffalo.
Correspondence concerning this article should be addressed to: Daniel F. Gros, Mental Health Service
116, Ralph H. Johnson VAMC, 109 Bee Street, Charleston, SC 29401; e-mail: [email protected]
JOURNAL OF CLINICAL PSYCHOLOGY, Vol . 67(3), 293--307 (2011) & 2010 Wiley Periodicals, Inc.
Published online in Wiley Online Library (wileyonlinelibrary.com). DOI : 10. 1002/ j cl p. 20763
heterogeneous diagnostic groups at rst, Brown and Barlow (2005) argued that such an
approach may address many of the short comings and sources of unreliability of the current
diagnostic system, including the current systems failure to convey severity of the majority of
disorders, classify overlapping symptoms between disorders, and identify symptoms that fall
below dened thresholds. In addition, Brown and Barlow suggested a hybrid approach to
diagnosis may be a necessary step to investigating the functionality of purely dimensional
models of psychopathology, which aim to eliminate the use of diagnostic categories.
Recently, Brown and Barlow (2009) took their proposal a step further by identifying
theoretically relevant dimensions for a hybrid model of the mood and anxiety disorders.
The authors suggested a possible framework for a dimensional classication system. Their
proposal comprised 10 candidate symptom dimensions: anxiety/neuroticism/behavioral
inhibition, behavioral activation/positive affect, unipolar depression, mania, somatic anxiety,
panic and related autonomic surges, intrusive cognitions, social evaluation, past trauma,
behavioral and interoceptive avoidance, and cognitive and emotional avoidance. Although
this model provides a framework of future hybrid models for the mood and anxiety disorders,
the authors acknowledged that the dimensions proposed in their model were limited to a
disparate array of clinician-administered and self-report measures and advocated for the
development and validation of a single multidimensional model and related measure.
If widely adopted, hybrid models may provide numerous advantages over the existing purely
categorical approaches to diagnoses (Brown & Barlow, 2009). Primarily, as noted above, the
replacement of overlapping diagnostic criteria with shared dimensional ratings across disorders
would improve the reliability of the diagnostic categories, thereby simplifying their diagnosis
and assessment. In addition, the identication of the shared dimensions underlying the mood
and anxiety disorders may provide a greater understanding of the biological, psychological, and
social factors associated with their development and maintenance. As much of the existing
literature has focused on identifying differences between the individual, overlapping disorders
(e.g., Barger & Sydeman, 2005; Moftt et al., 2007), hybrid models would shift the focus to
shared factors underlying all of the disorders. Similarly, a better understanding of common
factors among the disorders would inform the development of unied and/or transdiagnostic
treatments for sets of disorders (e.g., Ellard, Fairholme, Boisseau, Fachione, & Barlow, 2010),
rather than limiting treatment protocols to single disorders.
The goal of the present study was to investigate the general and specic symptom dimensions
associated with one specic emotional disordersocial phobiato inform the development of a
hybrid model of the disorder. A single disorder was chosen to focus the overwhelming large task
of developing a hybrid model across all of the mood and anxiety disorder. Social phobia was
specically chosen as it has been characterized by many of the same criticisms that have been
made against the current nosological system, including signicant comorbidity and hetero-
geneity (e.g., Antony & Rowa, 2005; Brown et al., 2001). If social phobia is conceptualized
within a hybrid model, it may be possible to separate the disorder into dimensional symptom
domains to assess both dimensions specic to social phobia (e.g., fear of embarrassment, social
skills decits) versus general dimensions that are relatively nonspecic across all of the anxiety
disorders (e.g., behavioral avoidance, physiological hyperarousal, general distress/neuroticism).
These dimensions may promote a rich description and understanding of the patterns of
symptoms and severities both within and across the diagnostic categories.
The two important steps in the development of a hybrid model for social phobia are to
(a) denitively identify the general and specic symptom dimensions associated with social
phobia, and (b) study the signicance and relative uniqueness of each proposed symptom
dimension. To that end, a broad range of dimensions were targeted in the present investigation.
These dimensions were identied by way of a thorough review of the literature, including the
theoretical descriptions (e.g., Antony & Rowa, 2005; Stein, Ono, Tajima, & Muller, 2004),
diagnostic features described in the DSM (APA, 2000), and empirical investigations (Brown,
Chorpita, & Barlow, 1998; Olivares, Garcia-Lopez, Hidalgo, & Caballo, 2004; Safren, Turk, &
Heimberg, 1998) of the symptoms relevant to social phobia and related disorders. This review
revealed several broad symptom dimensions that have been shown or hypothesized to be
relevant to social phobia specically or more broadly: (a) avoidance (e.g., behavioral,
294 Journal of Clinical Psychology, March 2011
interoceptive, cognitive avoidance, and safety behaviors; Brown, White, & Barlow, 2005; Magee
& Zinbarg, 2007; McManus, Sacadura, & Clark, 2008; Moitra, Herbert, & Forman, 2008); (b)
social skills decits (Stravynski & Amado, 2001); (c) physiological arousal (Edelmann & Baker,
2002); (d) impairment (Stein & Kean, 2000); (e) low positive affect (PA; Kashdan, 2007); and (f)
cognitive factors (i.e., anxious thoughts, insight, and anxiety sensitivity; Anderson & Hope,
2009; Rapee & Heimberg, 1997; Wakeeld, Horwitz, & Schmitz, 2005). The present
investigation was completed before Brown and Barlows (2009) proposed hybrid model was
published; however, the majority of the anxiety-relevant symptom dimensions from Brown and
Barlow (2009) also were represented in the indentied symptoms of the present investigation.
Although several of the proposed symptom dimensions have received mixed support in the
social phobia literature, the development of the list of proposed symptom dimensions was based
on the principles of substantive validity as outlined by Loevinger (1957) and rened by others
(Clark & Watson, 1995). Even the most modern psychometric methods are powerless to remedy
serious deciencies in the initial item pool. Thus, as Loevinger advised, the selection of the
proposed symptoms dimensions erred on the side of overinclusiveness, thereby incorporating
both dimensions with traditional status in the literature (e.g., DSM-dened symptoms) and
dimensions with more mixed evidence (e.g., thought avoidance or interoceptive avoidance).
Operational denitions for the complete list of proposed symptom dimensions are provided in
Table 1, along with an example of a citation that supports including each dimension.
To address the overinclusiveness of candidate symptoms and potential overlapping nature
of the symptoms, the present study incorporated a series of investigations designed to validate,
organize, and reduce these proposed symptom dimensions using a survey of experts and a
structural, principal components study. Together, the goal of these methods was to identify a
denitive list of specic and general dimensions to be incorporated into a hybrid model of
social phobia. We conducted two studies. In Study 1, we created operational denitions and a
large item pool for the symptom dimensions described in Table 1. The resulting denitions and
items were reviewed by a panel of experts in the treatment and research of the anxiety
disorders, involving both quantitative ratings and qualitative feedback, to revise and improve
denitions and items before data collection. In Study 2, we collected self-report ratings for the
revised item pool from a large unselected sample to investigate its structure and identify
empirically derived symptom dimensions to inform a hybrid model social phobia.
Study 1: Development of Initial Item Pool and Expert Review
In the rst study, a large item pool for each of proposed symptom dimension of social phobia
was developed. Similar to the selection of the proposed symptom dimensions, the development
of the initial item pool also was based on the principles of substantive validity (Clark &
Watson, 1995; Loevinger, 1957). Thus, the initial item pool was broader and more inclusive
than the dimension denitions presented earlier and erred on the side of overinclusiveness.
The rst investigation of the item pool was an expert review to improve content validity by
identifying and correcting any potential deciencies in the operational denitions and resulting
initial item poolsuch as confusing items or instructions, objectionable content, or missing
contentbefore collecting responses to the initial set of items (Haynes, Richard, & Kubany, 1995).
Method
Item pool. A pool of items was developed for each of the symptom domains described
earlier, including broad dimensions of avoidance (i.e., behavioral avoidance, interoceptive
avoidance, cognitive avoidance, and safety behaviors), social skills decits, low PA, functional
impairment, physiological arousal, and cognitive factors (i.e., anxious thoughts, insight, and
anxiety sensitivity). Unlike previous studies that relied on factor analyses of preexisting
measures of social anxiety (Olivares et al., 2004), the present study developed new items to
broadly and amply model the proposed symptom dimensions. In addition to the existing
measures that tapped similar constructs, item development also was based on a review of
published research and theory on social phobia and related symptoms. Operational denitions
295 Hybrid Model of Social Phobia
Table 1
Descriptions of Proposed Symptom Dimensions and Resulting Item Pool
Content domain Operational denition, example citation, and example items
Anxiety
sensitivity
Operational Denition: A belief that beyond any immediate physical discomfort, anxiety
and its accompanying symptoms may cause deleterious physical, psychological, or
social consequences
Sample Citation: Anderson and Hope (2009)
Sample Items: (1) I will eventually get so nervous that I will go crazy; (2) It worries me
that my heart races when I interact with people; (3) When my hands shake, I fear that
something is physically wrong with me
Anxious
thoughts
Operational Denition: Distorted or negative beliefs about a situation (e.g., being in the
grocery store will be unbearable), an individual (e.g., my boss will think that I am stupid),
or oneself (e.g., I will not be able to cope) that contribute to an individuals anxiety or fear
Sample Citation: Rapee and Heimberg (1997)
Sample Items: (1) If I share my ideas, they will be rejected; (2) I embarrass my friends in
social situations; (3) Most people would be right to ignore me
Behavioral
avoidance
Operational Denition: Avoidance of, or escape from an external situation involving the
possibility of social scrutiny or performance to reduce an individuals anxiety or
distress related to a specic situation
Sample Citation: Moitra, Herbert, and Forman (2008)
Sample Items: (1) I avoid speaking in front of groups of people; (2) I avoid eating in
public places; (3) I avoid talking to people in authority
Functional
impairment
Operational Denition: Impairment in ones daily routine, occupational (academic)
functioning, social activities, or relationships resulting from their avoidance or
anxious anticipation of feared social or performance situation
Sample Citation: Stein and Kean (2000)
Sample Items: (1) My anxiety has gotten in the way of my relationships; (2) My anxiety
has prevented me from taking certain jobs or promotions; (3) My anxiety has
prevented me from living a normal life
Insight Operational Denition: The recognition or lack of recognition of an individuals fear or
anxiety as excessive or unreasonable
Sample Citation: Wakeeld, Horwitz, and Schmitz (2005)
Sample Items: I realize that I worry more than I should; (2) If people told me that my
anxiety was unreasonable, I would agree with them; (3) I think that talking to a
professional about my anxiety might help me
Interoceptive
avoidance
Operational Denition: Avoidance of activities that directly produce symptoms of
physiological arousal. Examples include engaging in specic physical activities
(e.g., walking up a ight of stairs), consuming specic foods or drinks (e.g.,
drinking coffee), or visiting specic locations (e.g., sitting in a hot, stuffy room)
Sample Citation: Brown, White, and Barlow (2005)
Sample Items: (1) I avoid doing things that will make me sweat in front of others; (2)
I avoid situations that might make me blush or turn red in the face; (3) I avoid
physical activities that make my heart race
Lack of
positive affect
Operational Denition: A lack of pleasurable engagement with the environment,
involving fatigue, disinterest, lethargy, and a lack of enthusiasm and determination
Sample Citation: Kashdan (2007)
Sample Items: (1) I nd most activities to be boring or uninteresting; (2) I tend to be
unmotivated; (3) I rarely feel excited
Physiological
arousal
Operational Denition: Physiological hyperarousal (e.g., blushing, palpitations,
trembling, sweating) associated with the experience of anxiety or fear
Sample Citation: Edelmann and Baker (2002)
Sample Items: (1) My hands often shake when I speak to someone I do not know;
(2) My muscles tense up in social situations; (3) I get an upset stomach in social situations
Safety
behaviors
Operational Denition: Specic activities (e.g., monitoring ones speech, or over
preparing for a presentation), subtle avoidance (e.g., avoiding eye contact), or
checking behaviors (e.g., frequently checking ones appearance in the mirror)
designed to protect an individual from possible negative consequences associated with
feared situations or behaviors
296 Journal of Clinical Psychology, March 2011
were written for each symptom dimension and items were written by the research team
(i.e., seven PhD-level faculty in psychology, four graduate students in psychology, and six
undergraduate students in psychology across three universities), resulting in a large number of
items for each of the 11 proposed symptom dimensions (mean [M] number of items per
dimension 565.7; standard deviation [SD] 516). The resulting item pool was still too large to
effectively subject to an expert review; thus, the strongest items representing relatively unique
contents were selected in each dimension pool for use in the expert review (M539.7;
SD56.4). These selection decisions were based on the clarity and readability of the items (e.g.,
items with grammatical errors were removed) and the uniqueness of the proposed content
(e.g., only 12 items were retained to assess each specic symptom/construct). Item examples
for each proposed symptom dimension are provided in Table 1.
Expert review procedure. Next, an expert review study was conducted to investigate the
item pool. Researcher-practitioners in the eld of anxiety disorders were recruited to serve as
expert reviewers either in person at research conferences (e.g., annual meeting of the
Association for Behavioral and Cognitive Therapies) or by e-mail from members of the
research team. Each reviewer had (a) a PhD in clinical psychology and (b) advanced
experience in research and treatment of patients with anxiety disorders (total publications:
M564.5, SD553.2; anxiety-related publications: M545.7, SD543.7). To reduce the time
commitment associated with the review process, the full item pool was separated into three
questionnaires with subsets of the item pool and reviewers were paid $25 for each
questionnaire they completed. Expert reviewers were presented with thorough instructions
for completing the surveys and the complete list of proposed symptom dimensions,
operational denitions for each, and item pools for three or four symptoms dimensions.
Reviewers were asked to rate the relevance of each item to the proposed symptom
dimension and operational denition on a 5-point Likert scale, ranging from 0 (not at all
relevant) to 5 (very much relevant). In addition, qualitative feedback was collected from the
expert panel to assess the specic strengths and weaknesses of the symptom dimensions and
individual items (see below). The three separate surveys ranged from 128 to 190 total items,
incorporating both the quantitative and qualitative probes described above. Ten expert
reviewers completed 19 survey reviews (15 online; 4 paper-pencil), with each of the three
surveys reviewed by at least ve experts. Several incomplete surveys also were included by
reviewers (e.g., provided feedback on dimensions and/or operational denitions, but not the
individual items), resulting in a variable number of reviewers in the data below.
Table 1
Continued
Content domain Operational denition, example citation, and example items
Sample Citation: McManus, Sacadura, and Clark (2008)
Sample Items: (1) I need to know where the infrequently used restrooms are located; (2)
I wear dark clothing so that people will not notice me; (3) I always carry a bottle of
water with me
Social skills
decits
Operational Denition: The lack of requisite social skills (e.g., failure to make eye
contact, failure to convey warmth, failure to cooperate) to perform prociently in
various social situations
Sample Citation: Stravynski and Amado (2001)
Sample Items: (1) I think that my social skills are poor; (2) I do not know how to make
small talk; (3) I have trouble saying no others
Thought
avoidance
Operational Denition: An active resistance (e.g., thought suppression, mental repetition
of coping thoughts, distraction) to thoughts, particularly those that are believed to be
dangerous or threatening in some way
Sample Citation: Magee and Zinbarg (2007)
Sample Items: (1) I regularly try to block certain thoughts; (2) I try to avoid unwanted
thoughts in social situations; (3) I try to keep anxious thoughts out of my head
297 Hybrid Model of Social Phobia
Results and Revisions
Quantitative ndings. Descriptive statistics were computed for the relevance ratings on
the 436 items in the item pool. On average, the items were rated as much relevant on the
ve-point rating scale (M54.0; SD50.9), which suggests that the expert reviewers generally
approved of the relevance of the item pool as a whole. In fact, only 6 of 436 items received
average ratings of less than 3.0 or moderately relevant.
Qualitative Findings
Proposed symptom dimensions. Reviewers were asked, Is there signicant overlap
between any of the dimensions? If so, then which dimensions should be combined, erased,
and/or revised? In response, 6 of 12 reviews indicated that there was no signicant overlap in
the 11 content domains. The remaining reviewers raised concerns regarding the possible
overlap of behavioral avoidance and safety behaviors (4 of 12 reviews), social skills decits and
functional impairment (2 of 12 reviews), safety behaviors and thought avoidance (1 of 12
reviews), anxious thoughts and insight (1 of 12 reviews), and anxiety sensitivity and several of
the dimensions (2 of 12 reviews).
Reviewers also were asked, Are all of the dimensions relevant to social anxiety? If not,
which dimensions should be erased and/or revised? In response, several reviewers expressed
concern about the relevance of several content domains, including insight (5 of 13 reviews),
low PA (4 of 13 reviews), social skills decits (2 of 13 reviews), interoceptive avoidance (2 of 13
reviews), and anxiety sensitivity (1 of 13 reviews). In response to the two probesAre there
any signicant dimensions of social anxiety missing? If so, what symptom dimensions should
be added?, several new content domains were suggested, including self-focused attention
(2 of 12 reviews), critical thoughts (1 of 12 reviews), anticipatory or postevent anxiety (1 of 12
reviews), perfectionism (1 of 12 reviews), and higher order dimensions of personality, such as
neuroticism (1 of 12 reviews) and extraversion (1 of 12 reviews).
Operational denitions and item content. Reviewers were asked, Is the denition
complete and reasonable for the specied symptom dimension? If not, what should be
added and/or revised to improve the denition? In response, reviewers generally were
accepting of the operational denitions, noting only minor clarications regarding word
choice and reading level. The nal operational denitions are presented in Table 1. During
the review of the item pool, reviewers were given the opportunity to provide open feedback to
any/all of the items, resulting in additional feedback for 20.8% of the items. Feedback focused
upon the specic wording and grammar (50.5%), relevance to the proposed symptom
dimension and/or social anxiety (28.6%), or clarications or suggestions regarding the
intended content (20.9%).
Finally, reviewers were asked, Do you feel that these items were a good representation of
the target symptom dimension? If not, what content should be added and/or revised? If there
was specic missing content from the scale, please provide a few sample items to address this
concern. In response, 30 new items were recommended to address concerns of specic missing
content, including coping with substances, specic avoidant and interaction behaviors,
rumination, and perfectionism.
In response to the qualitative ndings, 16 new items were added to the item pool, many of
which were directly recommended by reviewers. In addition, to represent the new symptom
dimensions recommended by reviewers, items were identied within the item pool to match
the reviewers proposed symptom dimensions (e.g., self-focused attention, anticipatory or
postmortem anxiety, and perfectionism), and 14 new items were written to ensure that each
symptom dimension was represented by at least eight items. Revisions also were made to the
item pool based on the reviewers feedback on the items themselves: 5% of items were edited to
improve their grammar, wording, and readability and 3% of items were edited to improve
their relevance to social anxiety (e.g., adding in social situations).
298 Journal of Clinical Psychology, March 2011
Study 2: Exploring the Structure of the Item Pool
As discussed above, these initial dimensions were selected from theoretical discussions,
diagnostic criteria, and preliminary research, frequently involving the study and/or discussion
of a single construct, rather than the combined set of symptom dimensions. After the thorough
review by the panel of experts and resulting revisions to the item pool in Study 1, additional
empirical investigation was needed to adequately assess the overlapping constructs and overall
component structure. Thus, the goal of Study 2 was to collect self-report ratings for
all candidate items and investigate the structure of the item pool. Although the goal of the
present investigation was not focused on scale development, a psychometric evaluation of the
item pool, including a series of principal components analyses (PCAs), was used to identify an
optimal set of relatively distinct and homogeneous symptoms dimensions underlying the
sampled anxiety symptoms. Thus, we are using scale development procedures as a means of
theory and model development, rather than as a way to build measures per se (e.g., Cronbach
& Meehl, 1955; Clark & Watson, 1995; Loevinger, 1957; Simms & Watson, 2007; Simms,
Yuk, Thomas, & Simms, 2008).
Method
Participants. Responses were collected from an unselected sample (N5353) of
undergraduate students at a large northeastern university. The typical participant in the
sample was 19.0 years old (SD51.4), female (51%), and Caucasian (70.5% Caucasian, 14.2%
Asian, 5.9% Black, 2.5% Hispanic, and less than 1% Native American). Participants signed
up for the project through a research website setup by the psychology department and received
research course credit for their participation. No special exclusions or restrictions were used to
limit participation. The recommended guidelines for structural analyses were used to
determine the needed sample size: high expected communalities of the measured variables
(Z.6) and large variable-to-component ratios (Z30:1 ratio) allowed for a small to moderate
sample size in the present study (MacCallum, Widaman, Zhang, & Hong, 1999). Based on the
scores from previously validated measures of social phobia (see below), the sample
demonstrated above average symptoms of social phobia compared to community samples
(Orsillo, 2001). In addition, 14.3% of the sample met cutoff criteria for clinically signicant
social anxiety on the Social Interaction Anxiety Scale (SIAS; Gros, Hawk, & Moscovitch,
2010), suggesting that the sample adequately captures the appropriate range of anxiety
experiences.
Procedure
All procedures were approved by an appropriate institutional review board. Upon arrival
to the research lab, informed consent documents were reviewed and signed by all participants.
A questionnaire battery was administered using LabQ software (Simms, 2005) to participants
and included the revised 466-item pool, SIAS (Mattick & Clarke, 1998), Social Phobia
Scale (SPS; Mattick & Clarke, 1998), and Social Phobia Inventory (SPIN; Connor et al.,
2000), as well as several demographic questions. The SIAS, SPS, and SPIN were used to
detect random responding and investigate convergent validity of the resulting component
structure (see below). The items were administered in the same order to all participants.
Rigorous methods were included to detect and remove random and/or inconsistent responding
(see below).
Measures
Revised item pool. The item pool included 466 items, designed to assess a wide range of
general and specic symptoms of social phobia. The format of the items was based on other
popular measures of social anxiety (e.g., Mattick & Clarke, 1998), including a sentence format
with ve Likert-type responses, ranging from 0 (not at all characteristic or true of me) to 4
(extremely characteristic or true of me). The following instruction was provided for the
299 Hybrid Model of Social Phobia
items: For each question, please indicate the degree to which you feel the statement is
generally characteristic or true of you.
Social Interaction Anxiety Scale and Social Phobia Scale. The two measures were
developed to assess self-reported cognitive, affective, and behavioral reactions to general social
interactions (SIAS) and fear of situations that may involve the observation or evaluation by
others (SPS; Mattick & Clarke, 1998). Both measures comprise 20 questionnaire items and
each item is rated on a 5-point scale, ranging from 0 (not at all characteristic or true of me) to 4
(extremely characteristic or true of me). The measures have shown high internal consistency
(as range from .86 to .94) and test-retest reliability over 4 to 12 weeks (rs range from .66 to .93)
across a number of clinical, community, and student samples (for review, see Orsillo, 2001).
Social Phobia Inventory. The SPIN is a 17-item self-report measure that was designed
to assess fear, avoidance, and physiological arousal related to social phobia (Connor et al.,
2000). All items are rated on a ve-point scale, ranging from 0 (not at all) to 4 (extremely).
The SPIN has demonstrated high internal consistency (a4.82) and adequate test-retest
reliability (rs range from .78 to .89; Connor et al., 2000).
Data Reduction and Cleaning
The data were inspected for signicant missing responses, long strings of a single response
choice, extremely quick completers of the questionnaire battery, and random responders.
Ten participants were excluded from the analyses because of greater than 10 missing values
and an additional 30 participants were excluded for coding at least 100 consecutive items with
the same value. An analysis of the completion time of the questionnaire battery revealed no
outlying values; thus, no participants were excluded for completing the questionnaire too
quickly. To investigate the occurrence of random or problematically inconsistent responding,
10 items matched for content were selected from the item pool and related measures of social
anxiety, and absolute difference scores were computed and summed to create an index of
inconsistent responding. For example, item pool ]218, I avoid talking to people in
authority, and SPIN item 16, I avoid speaking to anyone in authority, represent one such
pair of items used to create the index. The values on the index then were investigated for
outlying values, dened as greater than three interquartile ranges above the median. Based on
these criteria, six participants were excluded because of inconsistent responding, resulting in a
nal sample of 307 participants. In participants with fewer than 10 missing values on the item
pool (2% of the total items), missing values were imputed with the mean of the top ve most
correlated responses in a participant-by-participant manner.
No differences in age, sex, or rst language (English vs. other) were found between the
excluded participants and the participants retained in the analyses. However, the rate of
exclusion and ethnic background were not independent, w
2
(5) 524.8, pso.01: Asian
Americans (22.0%) were excluded at a higher rate than Caucasians (10.4%), w
2
(1) 55.1,
po.05.
Results
Structural analyses. A series of PCAs was used to investigate the structure of
correlations among the measured variables and gather data to inform the item reduction
process (Fabrigar, Wegener, MacCallum, & Strahan, 1999).
1
A parallel analysis of eigenvalues
was used to establish the number of components to extract (Fabrigar et al., 1999). In the
1
Because our goal was to identify latent factors, we initially tried to factor analyze the item pool using
principal axis factor analytic procedures (Fabrigar et al., 1999). However, the software would not converge
on a solution, likely because of the large size of the correlation matrix.Thus, we opted instead to use
principal component analysis because of its computational simplicity, which resulted in interpretable
component solutions. Notably, these two variations of factor analysis typically result in highly similar
solutions, especially when the communalities among variables are high.
300 Journal of Clinical Psychology, March 2011
parallel analysis, the observed eigenvalue line crossed the random eigenvalue line after the
ninth component, suggesting that no more than nine components should be retained. Thus, we
examined the meaningfulness and interpretability of all solutions between two and eight
factors.
A series of eight obliquely rotated PCAs were conducted on the item pool to study its
structure and to develop empirically derived symptom dimensions. The ndings from these
eight PCAs are summarized in Table 2. Each of the models resulted in a slightly different
representation of the structure of the item pool. Based on the number of poorly loading items
(i.e., all loadingso.40) and cross-loading items (i.e., loadings4.30 on multiple components),
the less differentiated models (four components or fewer) demonstrated cleaner loadings.
Among the more differentiated models (ve components or more), the 7-component model
represented the cleanest of the models.
In addition to the number of poorly loading items, the pattern of loadings and
interpretability of the components also were investigated. The symptom dimensions were
identied by assigning clean loading items based on their highest component loading and
interpreted based on their originally targeted symptom dimensions and individual item
contents. In the less differentiated models, the 2-component and 3-component models
contained components comprised of heterogeneous items originating from multiple proposed
symptom dimensions and lacking centralized themes. The 4-component model was slightly
more interpretable and comprised components roughly consistent with depression (primary
items from low PA and anxious thoughts), interoceptive sensitivity (primary items from
interoceptive avoidance and anxiety sensitivity), and impairment (primary items from
functional impairment and insight). However, the largest component of the model included
numerous items from eight of the original eleven symptom dimensions, lacked a centralized
theme, and could not be interpreted. Thus, each of the less differentiated models was deemed
to provide a less ideal t to the data.
In the more differentiated models, the six components in the 7-component model were
distinctive, clearly interpretable, and psychologically meaningful. The content of these
components appeared to be consistent with the following labels: (I) social anxiety and
avoidance, (II) anhedonia, (III) functional impairment, (IV) interoceptive avoidance,
(V) thought avoidance, and (VI) coping with substances. The seventh component lacked any
clean loading items and thus was not interpreted. Exemplar items for each of the six components
were presented in Table 3. In contrast to the 7-component model, the 8-component model
created two components with highly overlapping content for physiological arousal and
Table 2
Summary Statistics for Principal Components Analyses for 466 Item Pool
Number of clean items
Model C1 C2 C3 C4 C5 C6 C7 C8 C9 o.40 CLs % PI
2-Component 184 149 66 67 28.5
3-Component 141 83 56 101 85 39.9
4-Component 121 41 41 47 139 77 46.4
5-Component 108 37 47 39 5 157 73 49.4
6-Component 103 46 34 51 7 0 145 80 48.3
7-Component 79 59 38 48 16 4 0 161 61 47.6
8-Component 78 43 51 23 16 15 4 2 166 68 50.2
9-Component 93 39 46 25 15 13 5 0 0 168 62 49.4
Note. Clean items were dened as those having a loading 4.40 on only one component and having cross-
loadings o.30 on all other components. C1C9 5components 1 to 9. o.40 5number of items with all
loadings below .40; CLs 5number of cross-loading items with components loadings of greater than .40 on
one component and greater than .30 on at least one more component; %PI 5total percent of problematic
items as categorized as o.40 and CL.
301 Hybrid Model of Social Phobia
avoidance with items of the two constructs intermixed within each component. The
8-component model also created a two-item avoidance of caffeine component. Similarly, the
9-component model also created two overlapping components for physiological arousal and
avoidance and two components without any cleanly loading items. The 6-component model
eliminated the coping with substances component, reduced the items in thought avoidance
component by half, and added many assorted items to the social anxiety and arousal
component, including several of the thought avoidance items. Thus, the 7-component model was
adopted, from which the rst six components served as the basis for provisional scales.
The items for each of the 11 proposed symptom dimensions were intermingled among the
six components of the 7-component model. Eight of 11 proposed symptom dimensions had at
least half of their items load onto a single component. A majority of the items from the
proposed interoceptive avoidance, low PA, and thought avoidance dimensions loaded on
similarly labeled components. Most items from the behavioral avoidance, physiological
arousal, and social skills decits dimensions loaded on the social anxiety and avoidance
component, representing a potentially broader construct. Most items from the functional
impairment and insight dimensions loaded on the functional impairment component.
However, the anxious thoughts, anxiety sensitivity, and safety behaviors dimensions
demonstrated balanced loadings across multiple components, suggesting that these constructs
represented a broad range of symptoms, rather than specic and unique constructs.
Relations to existing measures of social anxiety. As an exploratory evaluation of the
convergent validity of the symptom dimensions, the dimension scores were computed by
summing the clean loading items for each of the components, and correlations with the SIAS,
Table 3
Descriptions of Proposed Symptom Dimensions and Resulting Item Pool
Symptom dimension Top four loading items (abbreviated content)
Social anxiety and
avoidance
1. I avoid speaking in front of groups of people
2. I am unreasonably afraid of public speaking
3. I avoid performing in public (e.g., singing or dancing)
4. When I have to talk in front of others, I feel like I could choke
Functional impairment 1. I have a problem with my anxiety
2. If my anxiety keeps up, I will be alone
3. My social anxiety is out of control
4. My anxiety has prevented me from attending community events and/or
church services
Anhedonia 1. Very few things interest me
2. I do not nd people interesting to talk to
3. There is very little that gets me excited
4. I hardly ever feel interested in anything
Interoceptive avoidance 1. I avoid exercise that gets my blood pumping
2. I avoid eating hot foods that may make me look ushed in front of others
3. When I feel anxious in social situations, I fear that I might have a heart attack
4. I try not to wear heavy clothes in situations where I will probably feel nervous
Thought avoidance 1. I try not to get too wrapped up in thinking about my problems
2. If I am thinking about something I do not like, I immediately try to think of
something else
3. I try to push upsetting thoughts away
4. I try to replace my negative thoughts with positive thoughts
Coping with substances 1. I need to use alcohol and/or drugs when I go to parties or other social events
2. I need several drinks of alcohol in order to go to a bar or club
3. I often drink alcohol before a social event to help me manage
4. I have used drugs and/or alcohol to reduce my anxiety
302 Journal of Clinical Psychology, March 2011
SPS, and SPIN were computed. As presented in Table 4, a pattern of high correlations was
found between most dimensions and the alternative measures of social anxiety. The social
anxiety and avoidance (rs ranged from .82 to .86), anhedonia (rs ranged from .68 to .69),
functional impairment (rs ranged from .60 to .68), interoceptive avoidance (rs ranged from .64
to .72), and thought avoidance (rs ranged from .55 to .59) dimensions all demonstrated
moderate to large correlations with the SIAS, SPS, and SPIN. Coping with substances
evidenced the only difference, with signicantly smaller correlations with the three measures
(rs ranged from .25 to .29; Zs44.8; pso.01).
Discussion
The present set of studies identied a set of general and specic symptom dimensions
related to social phobia through an expert review and component analysis of an item pool that
was based upon a review of the theoretical (Stein et al., 2004), practical (Antony & Rowa,
2005), and empirical (Olivares et al., 2004) ndings in the social phobia literature. These
studies resulted in six symptom dimensions targeting social anxiety and avoidance,
interoceptive avoidance, thought avoidance, functional impairment, anhedonia, and coping
with substances.
Initially, 11 proposed symptom dimensions were identied through a review of the relevant
empirical and theoretical literature of social phobia. Several additional dimensions were added
to the initial list following the recommendations of the expert reviewers. However, through the
component analysis of the item pool, only six dimensions emerged in the nal model. Several
conclusions can be drawn from the resulting pattern of symptom dimensions.
First, the largest component in the model was characteristic of social anxiety and
avoidance. The component contained nearly one third of the clean loading items and a wide
range of symptoms, including a large number of items from the behavioral avoidance, safety
behaviors, and physiological arousal (in social interactions) initial dimensions. In addition, the
social anxiety and avoidance component demonstrated strongest correlations with the SIAS,
SPS, and SPIN. On a related note, the ndings for the large social anxiety and avoidance
factor are consistent with the content of the most common measures of social anxiety in the
current literature (Orsillo, 2001), focusing on the specic symptoms of social anxiety,
avoidance, and physiological arousal most associated with the disorder of social phobia,
rather than the content identied in the remaining components of the present study
(e.g., impairment, anhedonia, thought avoidance). Together, these ndings may suggest that
this large dimension of overlapping symptoms may be representative of the specic symptoms
of social phobia. More specically, if conceptualized within a hybrid model framework
(Brown & Barlow, 2005, 2009) involving both a specic diagnostic category (e.g., social
phobia) and several, more general symptom dimensions that cut across the mood and anxiety
Table 4
Correlations Among the Symptom Dimensions and With Other Measures of Social Anxiety
Scale No. a AIC 1 2 3 4 5 6 7 8
1. Social Anxiety and Avoidance 79 .99 .56
2. Functional Impairment 59 .99 .63 .83
3. Anhedonia 38 .97 .46 .78 .83
4. Interoceptive Avoidance 48 .97 .40 .76 .78 .74
5. Thought Avoidance 16 .92 .42 .70 .62 .53 .58
6. Coping with Substances 4 .87 .63 .34 .40 .43 .38 .29
7. Social Interaction Anxiety Scale 20 .91 .34 .86 .69 .68 .64 .55 .27
8. Social Phobia Scale 20 .94 .44 .82 .68 .60 .72 .59 .29 .81
9. Social Phobia Inventory 17 .92 .40 .83 .69 .61 .64 .55 .25 .83 .82
Note. N5307. All correlations are signicant, po0.01. no. 5number of items; AIC5average interitem
correlation.
303 Hybrid Model of Social Phobia
disorders, this dimension likely would represent the specic symptoms and/or diagnostic
category related to the disorder of social phobia.
Second, if the rst component of the model is representative of the specic symptoms of
social phobia, the remaining symptoms may represent nonspecic, general dimensions that are
shared to some extent with all of the anxiety disorders. These dimensions include interoceptive
avoidance, thought avoidance, functional impairment, anhedonia, and coping with
substances. Although the majority of these symptoms are more strongly associated with
other anxiety and mood disorders in the literature, each of these symptoms correlated
signicantly with the social anxiety and avoidance component and the existing scales of social
anxiety. In addition, there is empirical evidence for the connection of each of these constructs
and social phobia. For example, Brown et al. (1998) found that, unlike the other anxiety
disorders, both social phobia and depression were signicantly associated with low PA or
anhedonia in a large sample of outpatients with anxiety and depression. In addition, although
typically more associated with obsessive compulsive disorder (e.g., Purdon, Rowa, & Antony,
2005), recently, several authors have demonstrated signicant thought avoidance or
suppression in individuals with heightened social anxiety (e.g., Magee & Zinbarg, 2007).
Both functional impairment (DSM diagnostic criteria for most anxiety and mood disorders;
APA, 2000) and coping with substances (e.g., self-medication theory; Khantzian, 1985) also
have been linked to social phobia and other mood and anxiety disorders.
Third, when compared with the anxiety-relevant symptoms proposed in Brown and
Barlows (2009) dimensional classication system, several important patterns were identied.
A few of the unique symptom dimensions in Brown and Barlows model were directly
supported in the present ndings. Namely, behavioral activation/positive affect and cognitive
and emotional avoidance were consistent with the anhedonia and thought avoidance
components, respectively. Another proposed dimension, behavioral and interoceptive
avoidance, was only partially supported as the symptoms of interoceptive avoidance were
found to represent a separate, although highly correlated, component in the present ndings.
However, items consistent with the proposed dimensions of anxiety/neuroticism/behavioral
inhibition, somatic anxiety (initial dimension of physiological arousal), panic and related
autonomic surges (initial dimension of physiological arousal), intrusive cognitions (initial
dimension of anxious thoughts), social evaluation (initial dimension of behavioral avoidance),
and behavioral avoidance (initial dimension of behavioral avoidance) all loaded onto the same
factor of social anxiety and avoidance in the present ndings. As discussed above, this nding
may be complicated as Brown and Barlows model (2009) proposed these dimensions to
accompany a diagnostic category (e.g., social phobia); however, the PCAs in the present
investigation suggested that these dimensions did not represent separate constructs.
If replicated and expanded, the model identied in the present investigation could be used
to develop hybrid models for the anxiety disorders, involving disorder-specic categories
(e.g., social phobia, panic disorder, and posttraumatic stress disorder [PTSD]) and symptom
dimensions shared by each of the disorders (e.g., anhedonia, thought avoidance, functional
impairment, coping with substance, and interoceptive avoidance). Future research into hybrid
models should involve several steps. First, similar studies should be completed for each of the
anxiety disorders to identify the shared and unique symptoms of each (e.g., creation of a new
item pool, expert review of the proposed symptom domains and related items, and factor
analytic investigation of the item pool to determine symptom dimensions). Second, these
initial factor models should be replicated in various samples, including both clinical and
community populations. And third, each of the disorder-specic models should be combined
into a single factor analytic investigation in large diverse sample to identify the relatively
unique and shared dimensions across the anxiety disorders and create a single hybrid model
for the combine disorders. This new model would allow for the removal of these overlapping
symptoms from the disorder-specic diagnostic criteria (e.g., anhedonia symptoms from
PTSD), therefore reducing comorbidity and improving the diagnostic reliability (Watson,
2009). In addition, transdiagnostic treatments could be developed to treat the nonspecic
symptoms identied by the hybrid models for the anxiety disorders, simplifying the treatment
practices for these disorders (Ellard et al., 2010).
304 Journal of Clinical Psychology, March 2011
The present studies were limited in several respects that should be addressed in future
investigations of the hybrid models of social phobia and other related disorders. First, the
sample in Study 2 entirely comprised undergraduate students. As such, the sample was
younger and more highly educated than typical community adults. In addition, the college
samples are generally less symptomatic than clinical samples on measures of anxiety (e.g.,
Orsillo, 2001). Although the present study sought to assess a wide range of anxiety
symptomatology, rather than limiting the investigation to clinical presentations of social
phobia, discrepancies between the present samples and community samples were still present.
A second limitation of the present studies is the focus on social phobia, rather than a
comprehensive investigation across multiple types of psychopathology. Although social
phobia may be representative of the anxiety disorders in a variety of ways (e.g., high
comorbidity, heterogeneity), it is important that future research incorporate alternative
presentations of the anxiety to further investigate the hypotheses proposed by the present
study regarding the specic (social anxiety and avoidance) and general (interoceptive
avoidance, thought avoidance, functional impairment, anhedonia, and coping with
substances) symptoms of social phobia. And, third, the component structure identied in
the present investigation relied more heavily on theoretical models than empirical ndings,
suggesting that replication studies are needed to further validate the proposed factor structure.
In conclusion, the present studies investigated symptoms associated with social phobia in
effort to develop a hybrid model of the disorder. These analyses lead to the identication of six
dimensions of social phobia: social anxiety and avoidance, interoceptive avoidance, functional
impairment, anhedonia, thought avoidance, and coping with substances. Although these
ndings are preliminary and require replication of the factor structure, they provide an initial
foundation on which future studies of hybrid models of social phobia and/or other anxiety
disorders should be based.
References
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed.
text revision). Washington, DC: Author.
Anderson, E.R., & Hope, D.A. (2009). The relationship among social phobia, objective and perceived
physiological reactivity, and anxiety sensitivity in an adolescent population. Journal of Anxiety
Disorders, 23, 1826.
Antony, M.M., & Rowa, K. (2005). Evidence-based assessment of anxiety disorders in adults.
Psychological Assessment, 17, 256266.
Barger, S.D., & Sydeman, S.J. (2005). Does generalized anxiety disorder predict coronary heart disease
risk factors independently of major depressive disorder? Journal of Affective Disorders, 88, 8791.
Brown, T.A., & Barlow, D.H. (2005). Dimensional versus categorical classication of mental disorders in
the fth edition of the Diagnostic and Statistical Manual of Mental Disorders and beyond: Comment
on the special section. Journal of Abnormal Psychology, 114, 551556.
Brown, T.A., & Barlow, D.H. (2009). A proposal for a dimensional classication system based on the
shared features of the DSM-IV anxiety and mood disorders: Implications for assessment and
treatment. Psychological Assessment, 21, 256271.
Brown, T.A., Campbell, L.A., Lehman, C.L., Grisham, J.R., & Mancill, R.B. (2001). Current and lifetime
comorbidity of the DSM-IV anxiety and mood disorders in a large clinical sample. Journal of
Abnormal Psychology, 110, 585599.
Brown, T.A., Chorpita, B.F., & Barlow, D.H. (1998). Structural relationships among dimensions of the
DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic
arousal. Journal of Abnormal Psychology, 107, 179192.
Brown, T.A., White, K.S., & Barlow, D.H. (2005). A psychometric reanalysis of the Albany Panic and
Phobia Questionnaire. Behaviour Research and Therapy, 43, 337355.
Clark, L.A., & Watson, D. (1995). Constructing validity: Basic issues in objective scale development.
Psychological Assessment, 7, 309319.
305 Hybrid Model of Social Phobia
Connor, K.M., Davidson, J.R.T., Churchill, L.E., Sherwood, A., Foa, E.B., & Wesler, R.H. (2000).
Psychometric properties of the Social Phobia Inventory (SPIN). British Journal of Psychiatry, 176,
379386.
Cronbach, L.J., & Meehl, P.E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52,
281302.
Edelmann, R.J., & Baker, S.R. (2002). Self-reported and actual physiological responses in social phobia.
British Journal of Clinical Psychology, 41, 114.
Ellard, K.K., Fairholme, C.P., Boisseau, C.L., Fachione, T.J., & Barlow, D.H. (2010). Unied protocol
for the transdiagnostic treatment of emotional disorders: Protocol development and initial outcome
data. Cognitive and Behavioral Practice, 17, 88101.
Fabrigar, L.R., Wegener, D.T., MacCallum, R.C., & Strahan, E.J. (1999). Evaluating the use of
exploratory factor analysis in psychological research. Psychological Methods, 4, 272299.
Gros, D.F., Hawk, Jr., L.W., & Moscovitch, D.A. (2010). The psychophysiology of social anxiety:
Emotional modulation of the startle reex during socially-relevant and irrelevant pictures.
International Journal of Psychophysiology, 73, 207211.
Haynes, S.N., Richard, D.C.S., & Kubany, E.S. (1995). Content validity in psychological assessment:
A functional approach to concepts and methods. Psychological Assessment, 7, 238247.
Kashdan, T.B. (2007). Social anxiety spectrum and diminished positive experiences: Theoretical synthesis
and meta-analysis. Clinical Psychology Review, 27, 348365.
Khantzian, E.J. (1985). The self-medication hypothesis of addictive disorders. American Journals of
Psychiatry, 142, 12591264.
Krueger, R.F., Watson, D., & Barlow, D.H. (2005). Introduction to the special section: Toward a
dimensionally based taxonomy of psychopathology. Journal of Abnormal Psychology, 114, 491493.
Loevinger, J. (1957). Objective tests as instruments of psychological theory. Psychological Reports, 3, 635694.
MacCallum, R.C., Widaman, K.F., Zhang, S., & Hong, S. (1999). Sample size in factor analysis.
Psychological Methods, 4, 8499.
Magee, J.C., & Zinbarg, R.E. (2007). Suppressing and focusing on a negative memory in social anxiety:
Effects on unwanted thoughts and mood. Behavioral Research and Therapy, 45, 28362849.
Mattick, R.P., & Clarke, J.C. (1998). Development and validation of measures of social phobia scrutiny
fear and social interaction anxiety. Behaviour Research and Therapy, 36, 455470.
McManus, F., Sacadura, C., & Clark, D.M. (2008). Why social anxiety persists: An experimental
investigation of the role of safety behaviours as a maintaining factor. Journal of Behavior Therapy and
Experimental Psychiatry, 39, 147161.
Mineka, S., Watson, D., & Clark, L.A. (1998). Comorbidity of anxiety and unipolar mood disorders.
Annual Review of Psychology, 49, 377412.
Moftt, T.E., Caspi, A., Harrington, H., Milne, B.J., Melchior, M., Goldberg, D., et al. (2007).
Generalized anxiety disorder and depression: Childhood risk factors in a birth cohort followed to
age 32. Psychological Medicine, 37, 441452.
Moitra, E., Herbert, J.D., & Forman, E.M. (2008). Behavioral avoidance mediates the relationship
between anxiety and depressive symptoms among social anxiety disorder patients. Journal of Anxiety
Disorders, 22, 12051213.
Olivares, J., Garcia-Lopez, L.J., Hidalgo, M.D., & Caballo, V. (2004). Relationship among social anxiety
measures and their invariance: A conrmatory factor analysis. European Journal of Psychological
Assessment, 20, 172179.
Orsillo, S.M. (2001). Measures for social phobia. In M.M. Antony, S.M. Orsillo, & L. Roemer (Eds.),
Practitioners guide to empirically based measures of anxiety. New York: Springer.
Purdon, C., Rowa, K., & Antony, M.M. (2005). Thought suppression and its effects on thought frequency,
appraisal and mood state in individuals with obsessive-compulsive disorder. Journal of Anxiety
Disorders, 43, 93108.
Rapee, R.M., & Heimberg, R.G. (1997). A cognitive-behavioral model of anxiety in social phobia.
Behaviour Research and Therapy, 35, 741756.
Safren, S.A., Turk, C.L., & Heimberg, R.G. (1998). Factor structure of the Social Interaction Anxiety
Scale and the Social Phobia Scale. Behaviour Research and Therapy, 36, 443453.
Simms, L.J. (2005). LabQ computerized test administration software (Version 0.1b) [Computer Software].
Buffalo, NY: Author.
306 Journal of Clinical Psychology, March 2011
Simms, L.J., & Watson, D. (2007). The construct validation approach to personality scale construction.
In R. Robins, C. Fraley, & R. Krueger (Eds.), Handbook of research methods in personality
psychology. Guilford Press.
Simms, L.J., Yuk, T., & Thomas, J.P., & Simms, E.N. (2008). Exploring evaluative person descriptors
through scale development. Journal of Research in Personality, 42, 12711284.
Stein, M.B., & Kean, Y. (2000). Disability and quality of life in social phobia: Epidemiologic ndings.
American Journal of Psychiatry, 157, 16061613.
Stein, D.J., Ono, Y., Tajima, O., & Muller, J.E. (2004). The social anxiety disorder spectrum. Journal of
Clinical Psychiatry, 65, 2733.
Stravynski, A., & Amado, D. (2001). Social phobia as a decit in social skills. In S.G. Hofmann, &
P.M. DiBartolo (Eds.), From social anxiety to social phobia: Multiple perspectives (pp. 107129).
Needham Heights: Allyn and Bacon.
Wakeeld, J.C., Horwitz, A.V., & Schmitz, M.F. (2005). Are we overpathologizing the socially anxious?
Social phobia from a harmful dysfunction perspective. Canadian Journal of Psychiatry, 50, 317319.
Watson, D. (2005). Rethinking the mood and anxiety disorders: A quantitative hierarchical model for
DSM-V. Journal of Abnormal Psychology, 114, 522536.
Watson, D. (2009). Differentiating the mood and anxiety disorders: A quadripartite model. Annual
Review of Clinical Psychology, 5, 221247.
Widiger, T.A., & Samuel, D.B. (2005). Diagnostic categories or dimensions? A Question for the
Diagnostic and Statistical Manual of Mental Disorders (5th ed.), Journal of Abnormal Psychology,
114, 494500.
307 Hybrid Model of Social Phobia