Problematic Shopping Behavior: An Item Response Theory Examination of the Seven-Item Bergen Shopping Addiction Scale

Abstract

There has been an increasing amount of research examining problematic shopping behavior (PSB), often referred to in the psychological literature as “compulsive buying” or “shopping addiction.” A popular scale for assessing the risk of PSB is the seven-item Bergen Shopping Addiction Scale (BSAS). To expand our knowledge of the psychometric properties of this instrument, the present study employed Item Response Theory (IRT) and differential item functioning analyses (DIF) while concurrently attempting to determine a preliminary cut-off point. A relatively large community sample completed the BSAS online (N = 968, Mage = 29.5 years, SDage = 9.36, 32.5% women). IRT analyses showed differences regarding the BSAS items’ discrimination, difficulty, and precision, with a raw score exceeding 23 (out of 28) indicating a higher risk of shopping addiction. Finally, while most BSAS items operated equally among males and females, Item 2 (mood modification) required a higher level of shopping addiction behaviors to be endorsed by males. The BSAS functions as a reliable assessment of the risk of shopping addiction, particularly between average and high levels of the trait. Clinical implications are discussed in light of these findings.

Psychometric properties of the Bergen Social Media Addiction Scale: An analysis using item response theory

“…In line with previous studies, questions addressing relapse (i.e., inability to cut down) assume the awareness of problematic behaviors and intent of cutting down, yet the limited reliability observed here may suggests that individuals within 0SD to +2SD in PSMU may not perceive their social media use as problematic (i.e. lacking reflection), and thus not intent to reduce it ( Zarate, Fullwood, Prokofieva, Griffiths, and Stavropoulos, 2022 ). Additionally, none of the items provided sufficient information to reliably identify individuals with significantly low levels of PSMU (−3SD to −2SD).…”

Exploratory structural equation modeling: a streamlined step by step approach using the R Project software

“…Using these steps, a five-factor ESEM was fitted with all non-primary items approximating zero (~ 0). This model showed acceptable fit indices (χ 2 [185] = 1372.931, p < 0.001, CFI = 0.952, TLI = 0.922, RMSEA = 0.041), most items loaded significantly on all five latent factors and latent factors covaried (see Fig. 4).…”

“…Human behavior researchers commonly employ selfreport scales to assess individuals' experiences, including complex psychiatric presentations [1,2]. Such instruments usually include items or indicators assumed to capture latent constructs [3].…”

Online behavioral addictions: Longitudinal network analysis and invariance across men and women.

“…Previous literature has identified gender differences in an individual's propensity to develop certain types of addictive behaviors (Charzyńska et al, 2021;Chen et al, 2017;McHugh et al, 2018;Pelissier & Jones, 2005;Pontes et al, 2019;Tang & Koh, 2017). For example, while males are more vulnerable to developing substance, gaming, gambling, and online porn consumption-related addictions, females appear more likely to experience problematic behaviors associated with excessive use of social networking sites, shopping, and even studying (Charzyńska et al, 2021;McHugh et al, 2018;Pelissier & Jones, 2005;Pontes et al, 2019;Zarate, Fullwood, et al, 2022). Similarly, research has indicated that there may be gender differences in the age of onset of such problematic behaviors, especially for behavioral addictions (e.g., gambling), with females more likely to have a mid-to-later life onset and males commonly displaying earlier difficulties (APA, 2013).…”

“…Thus, Mood modification to avoid distress or social isolation could increase the likelihood of transitioning into other problematic online behaviors. Moreover, studies reported that individuals might develop problematic behaviors such as online gaming (Gomez et al, 2018), problematic online shopping (Zarate, Fullwood, et al, 2022), or disordered online gambling (Stavropoulos, Monger, et al, 2022) to "escape" their problems (i.e., Mood modification). Finally, the results presented here support the hypothesized "digital convergence", suggesting that certain features of online platforms may facilitate further problematic online behaviors (Delfabbro & King, 2020).…”