# Evaluating the psychometric properties of the Generic Conspiracist Beliefs Scale: a Rasch modelling approach

## TL;DR

This paper evaluates the psychometric properties of the Generic Conspiracist Beliefs Scale using Rasch analysis to improve its reliability and validity.

## Contribution

The study applies Rasch analysis to the GCBS, offering a more rigorous evaluation than traditional methods.

## Key findings

- The GCBS supports a five-factor structure and shows satisfactory reliability and item/person fit.
- Items displayed invariance across age groups, but the Control of Information subscale had lower reliability.
- Strong associations were found between Government Malfeasance and Control of Information dimensions.

## Abstract

Within psychological research, the Generic Conspiracist Beliefs Scale (GCBS) has emerged as the predominant measure of conspiratorial ideation. Although studies report that the GCBS is psychometrically satisfactory (i.e., valid and reliable) and possesses a robust five-factor multidimensional latent structure, analysis has employed traditional measurement evaluation. Acknowledging this, the present paper evaluated the GCBS using Rasch analysis. Rasch analysis was necessary because it provides a more rigorous examination of measurement properties than classical test theory. This allows theorists to refine scales, develop precise measures, and advance the quality and comparability of findings across studies.

The study sample comprised 2,987 UK participants (Mage = 32.13, SD = 14.18), with 1,335 males and 1,652 females. Data were collected via online survey and subjected to a series of psychometric tests to determine dimensionality and item-level performance.

Preliminary evaluation confirmed the presence of more than one dimension. Accordingly, the authors proceeded with multidimensional Rasch modelling, which verified that data supported the original five-factor structure. Moreover, the GCBS demonstrated satisfactory item/person fit and reliability, and items mostly displayed invariance across age (young, 18–24 years vs. older adults, 25–88 years). There were strong positive associations within GCBS dimensions; the highest was between Government Malfeasance (i.e., illegal or unethical state actions) and Control of Information (i.e., the deliberate manipulation or suppression of truth).

The presence of a multidimensional, correlated latent structure supported the notion that the GCBS assesses a range of construct breadth. Future research should evaluate GCBS item functioning across heterogeneous samples, specifically non-Western, non-English speaking, and different educational groups. Furthermore, since the Control of Information subscale demonstrated lower reliability, further refinement is essential for its effective use as a distinct index of conspiratorial ideation.