Social anxiety questionnaire (SAQ): Development and preliminary validation

Highlights

• A new, theory-driven instrument for assessing social anxiety is proposed.

• A theoretical framework is provided by Clark and Wells’ model of social anxiety.

• The 5-factor questionnaire was developed and validated across 4 samples.

• The questionnaire is reliable, and both theoretically and empirically valid.

Abstract

Background

The Social Anxiety Questionnaire (SAQ) was designed to assess five dimensions of social anxiety as posited by the Clark and Wells' (1995; Clark, 2001) cognitive model.

Methods

The development of the SAQ involved generation of an item pool, followed by a verification of content validity and the theorized factor structure (Study 1). The final version of the SAQ was then assessed for reliability, temporal stability (test re-test reliability), and construct, criterion-related, and contrasted-group validity (Study 2, 3, and 4).

Results

Following a systematic process, the results provide support for the SAQ as reliable, and both theoretically and empirically valid measure. A five-factor structure of the SAQ verified and replicated through confirmatory factor analyses reflect five dimensions of social anxiety: negative self-processing; self-focused attention and self-monitoring; safety behaviours; somatic and cognitive symptoms; and anticipatory and post-event rumination.

Limitations

Results suggest that the SAQ possesses good psychometric properties, while recognizing that additional validation is a required future research direction. It is important to replicate these findings in diverse populations, including a large clinical sample.

Conclusions

The SAQ is a promising measure that supports social anxiety as a multidimensional construct, and the foundational role of self-focused cognitive processes in generation and maintenance of social anxiety symptoms. The findings make a significant contribution to the literature, moreover, the SAQ is a first instrument that offers to assess all, proposed by the Clark–Wells model, specific cognitive-affective, physiological, attitudinal, and attention processes related to social anxiety.

Introduction

Social anxiety refers to fear of social situations due to concerns about being scrutinised, judged or embarrassed, including anxiety over social interactions, with excessive levels considered to constitute a disorder, called social anxiety disorder (SAD) in the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5; American Psychiatric Association, 2013). With the prevalence of SAD (see Heimberg et al., 2000, Kessler et al., 2005, Stein et al., 2017), and both theoretical and empirical interest in social anxiety growing, there is a need for measures of social anxiety that are brief and can be used more practically while assessing the construct comprehensively.

A review of currently available and used measures of social anxiety, for instance, the Liebowitz Social Anxiety Scale (LSAS; Liebowitz, 1987), the Social Phobia Weekly Summary Scale (SPWSS; Clark et al., 2003), the Social Interaction Phobia Scale (Carleton et al., 2009, Reilly et al., 2012), the Social Interaction Anxiety Scale (SIAS) and the Social Phobia Scale (SPS; Mattick and Clarke, 1998), the Social Cognitions Questionnaire (SCQ; Wells et al., 1993), and the Social Phobia Inventory (SPIN; Connor et al., 2000, Carleton et al., 2010), revealed some issues to be resolved. First and foremost, there exists no Polish language reliable and valid self-report measures of social anxiety. For instance, researchers in Poland have used either a translated version of the LSAS without empirical validation (c.f., Dabkowska, 2007, Dabkowska, 2008) or modified version of the STAI (State-Trait Anxiety Inventory; Spielberger et al., 1983; c.f., Karas, 2005) as a measure of social anxiety. The development of appropriate, valid, and reliable assessment tool of social anxiety is necessary for further studies (e.g., in Poland, research on social anxiety are still relatively rare, probably due to the insufficient availability of valid measures). Moreover, considering the nature of social anxiety (APA, 2013; see also Bögels et al., 2010, Heimberg et al., 2014b), the assessment should not only include the shared features of fear, avoidance, and anxiety, which have been reflected by most measures (c.f., Carleton et al., 2009, Connor et al., 2000, Liebowitz, 1987, Mattick and Clarke, 1998), but also unique features of social anxiety, such as specific cognitive-affective, physiological, attitudinal, behavioural, and attention performance processes should be addressed. Few current assessment instruments of social anxiety capture cognitive schemas, but they are however only narrowed down to the facet (c.f., Wells et al., 1993). Thus, the measures available for the assessment of social anxiety are often limited to avoidance ratings or thoughts and feelings and do not fully cover the breadth of social phobia domains. In contrast, indeed the six-item SPWSS (Clark et al., 2003,2006) comprises ratings of social anxiety, avoidance, self-focused attention, anticipatory processing and post-event rumination, nevertheless the scale was developed without empirical validation and psychometric evaluation. To address the above-existed issues; including the multidimensionality of SAD (Clark, 2001; see also Heimberg et al., 2014a, Heimberg et al., 2014b), and, importantly, a need for theoretically and psychometrically sound measures of social anxiety in Poland; a new instrument that captures specific cognitive-affective, physiological, attitudinal, and attention performance processes related to social anxiety was proposed and developed. An access to theoretically and psychometrically valid assessment instruments is important to conduct research and provide quality patient care.

The newly developed instrument, called the Social Anxiety Questionnaire (SAQ), draws upon one of the most prominent, well-studied, and fruitful models providing an evidence-based way to conceptualize and treat SAD – Clark and Wells, 1995, Clark, 2001) cognitive model of social anxiety. The model, unique in the synthesis of the writings of earlier theorists (e.g., Beck et al., 1985, Heimberg and Barlow, 1988, Leary, 1983), is largely focused on the maintaining processes of SAD and attempts to explain why individuals with social phobia fail to benefit from the naturalistic exposure that is provided by their everyday interactions with other people. The four maintenance processes that are highlighted in the model are: (1) increased self-focused attention and a linked decrease in observation of other people and their responses (detailed monitoring and observation of oneself); (2) use of misleading internal information (images and feelings) to make excessively negative inferences about how one appears to others (generate a negative view of the self as a social object); (3) extensive use of overt and covert safety-seeking behaviours to prevent or minimize the feared social catastrophes (e.g., attempts to avoid being the centre of attention, avoiding eye contact, self-monitoring, not talking about oneself); and (4) problematic pre- and post-event processing (i.e., engaging in negatively biased anticipatory processing before entering feared social situations and prolonged, negatively biased, post-event processing). These processes are accompanied by marked arousal and fear. Persons with SAD are particularly concerned about somatic and cognitive symptoms of anxiety (e.g., tremor, sweating, feeling hot in the face, and mental blanks) that they think could be observed by others (Clark, 2001; see also Heimberg et al., 2014b). Because of the perceived significance of the symptoms, they are often hypervigilant and interpret such symptoms as signs of impending or actual failure to meet their desired standards of social interactions. Additionally, the hypervigilance and a variety of safety behaviours often increase the visibility and the subjective intensity of the somatic and cognitive symptoms, driving the vicious cycle (for more details, see Clark and Wells, 1995, Clark, 2001).

Evidence supporting the Clark and Wells’ model has been found in research conducted in child, adolescent, and adult populations, using both correlational and experimental designs (c.f., Clark, 2001, Clark, 2004). Each specific aspect of the model has received substantial empirical support. For example, studies have shown that socially anxious individuals display higher levels of self-focused attention (Hodson et al., 2008, Miers et al., 2013, Pineles and Mineka, 2005, Rapee and Abbott, 2007, Spurr and Stopa, 2002), evaluate social events more negatively (Alfano et al., 2006, Inderbitzen-Nolan et al., 2007, Norton and Hope, 2001, Stopa and Clark, 2000), report negative self-related thoughts (Dodge et al., 1988, Ranta et al., 2014) and negative observer-perspective images of self in social situations (Hackmann et al., 1998, Wells et al., 1998). Furthermore, individuals with high levels of social anxiety reported using a greater number of safety behaviours, more frequently, and in a greater number of situations than those with low levels of social anxiety (Hodson et al., 2008, McManus et al., 2008). Holding negative self-images in mind during social situations has been shown to generate higher levels of fear and anxiety, stronger beliefs that one's anxiety symptoms are visible, and reduced ratings of subjective performance in social-evaluative situations (Hirsch et al., 2003, Spurr and Stopa, 2003). There is also a substantial body of empirical evidence demonstrating that SAD individuals engage in self-focused cognition prior, during, and following social situations, generating a cycle of dysfunctional thinking that contributes to the maintenance of social anxiety (for review, see Norton and Abbott, 2016). The findings provide support for all specific aspects of the Clark-Wells model underscoring the multidimensionality of SAD, which constitutes the theoretical foundation for a novel measure of social anxiety symptoms.

The current research aimed to, firstly, develop the Social Anxiety Questionnaire (SAQ), as a new measure of social anxiety symptoms that beyond assessing anxiety, avoidance, and distress, reflects negative social cognitions, safety behaviours, self-focused attention, and pre- and post-event processing – as posited by Clark and Wells, 1995, Clark, 2001) model of SAD; and secondly, validate the SAQ in Polish samples. Moreover, the additional goal was to create a questionnaire, which captures all of the factors and is not being overly long. A questionnaire length is one of the important factors in determining response burden manifested in lower response rate, reduced completion, and reduced data quality (Rolstad et al., 2011). Overall, using four samples recruited from clinical and nonclinical settings, the present research set out to assess the psychometric properties of the SAQ as a novel measure of social anxiety symptoms. Two studies were conducted to verify and replicate the hypothesized factor structure of the SAQ (Study 1 and Study 2). Thereafter, internal consistency, test re-test reliability, and convergent and discriminant validity were assessed (see Study 2, 3, and 4). This sequence of psychometric evaluations is intended to provide an initial assessment of the SAQ, while recognizing that additional validation will be a required future research direction.