MAIN The Covert and Overt Reassurance Seeking Inventory (CORSI): Development, validation and psychometric analyses Adam S. Radomsky* , Rachael L. Neal , Chris L. Parrish, Stefanie L. Lavoie and Sarah E. Schell Concordia University, Montréal, Quebec, Canada *Corresponding author. Email: Adam.Radomsky@concordia.ca (Received 18 December 2018; revised 5 August 2020; accepted 7 August 2020) Abstract Background: Reassurance seeking (RS) is motivated by perceived general and social/relational threats across disorders, yet is often under-recognized because it occurs in covert (i.e. subtle) and overt forms. Covert safety-seeking behaviour may maintain disorders by preventing corrective learning and is therefore important to identify effectively. Aims: This study presents the validation and psychometric analyses of a novel measure of covert and overt, general and social/relational threat-related interpersonal RS. Method: An initial 30-item measure was administered to an undergraduate sample (N = 1626), as well as to samples of individuals diagnosed with obsessive compulsive disorder (OCD; n = 50), anxiety disorders (n = 60) and depression (n = 30). The data were subjected to exploratory and confirmatory factor analyses, and validation analyses. Results: An exploratory factor analysis using principal axis factoring with oblique rotation yielded five interpretable factors, after removing four complex items. The resulting 26-item measure, the Covert and Overt Reassurance Seeking Inventory (CORSI), evidenced good convergent and divergent validity and accounted for 54.99% of the total variance after extraction. Factor correlations ranged from r = .268 to .736, suggesting that they may be tapping into unique facets of RS behaviour. In comparison with undergraduate participants, all clinical groups had significantly higher total scores [t (51.80–840) = 3.92–5.84, p < .001]. The CFA confirmed the five-factor model with good fit following the addition of four covariance terms (goodness of fit index = .897, comparative fit index = .918, Tucker–Lewis index = .907, root mean square error approximation = .061). Conclusion: The CORSI is a brief, yet comprehensive and psychometrically strong measure of problematic RS. With further validation, the CORSI has potential for use within clinical and research contexts. Keywords: anxiety; assessment; depression; obsessive-compulsive disorder or OCD; reassurance seeking; validation Introduction Reassurance seeking (RS) is a behaviour recognized for its role in maintaining various mental disorders. Despite its prevalence (e.g. Starcevic et al., 2012), facets of RS are under-represented in existing measures. The aim of this paper is to present the development, validation and psychometric analyses of a novel measure of RS behaviour. Individuals with mental disorders including obsessive-compulsive disorder (OCD), depression and anxiety disorders [e.g. generalized anxiety disorder (GAD); social anxiety disorder (SAD)] may seek reassurance in ways that become repetitive and detrimental (e.g. American Psychiatric Association, 2013; Beesdo-Baum et al., 2012; Coyne, 1976; Heerey and Kring, 2007; Parrish and Radomsky, 2010; Warwick and Salkovskis, 1985). Individuals with such © British Association for Behavioural and Cognitive Psychotherapies 2020 Behavioural and Cognitive Psychotherapy, 2020, page 1 of 18 doi:10.1017/S1352465820000703