Competitive Latent Structures for the Comic Style Markers: Developing a Psychometrically Sound Short Version Using Spanish and US American Samples

Abstract

The Comic Style Markers (CSM) is a questionnaire that allows a fine-grained description of how people differ in the way they display humor in their daily lives. It includes 48 statements capturing eight interrelated, yet distinct comic styles: fun, irony, wit, sarcasm, benevolent humor, satire, nonsense humor, and cynicism. Despite the independent conceptual roots of these humorous domains, the analysis of the CSM scales’ latent structure shows that their empirical distinction needs to be improved. Using the information derived from a competitive latent approach, including confirmatory factor analysis, bifactor analysis, and exploratory structural equation modeling, we proposed and validated a shorter 24-item version of the CSM in a large sample of 925 Spanish individuals (SP-CSM-24). This scale-refinement improved the psychometric differentiation of the eight comic styles without undermining the good internal consistency and the temporal stability of the CSM scores. Strong invariance was held for gender and age groups, and partial scalar invariance for countries also emerged using a sample of 318 U.S. American adults. Structural equation modeling also corroborated a convincing test-criterion validity for the SP-CSM-24, with dispositional expressions of benevolent humor (positively) and cynicism (negatively) outperforming other comic styles in accounting for individuals’ well-being.

Declaration of interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Notes

1 It has been stated that when RMSEA for the null model is lower than .158, incremental fit indices are less informative (Kenny et al., 2015; Moreira & Inman, 2021). The RMSEA for the null model in this research was .132.

2 U.S. Americans also had higher scores on the sarcastic comic style than Spaniards; however, we suggest interpreting this comparison with caution as 2 items of this domain were found to be non-invariant.

3 Domain-level statistic descriptives, internal consistencies (all αs/ ω_ts ≥ .80/81), and interrelations of the well-being measures are presented in Table SM9 in the Supplementary Materials.

Additional information

Funding

This study was supported by the FEDER/Junta de Andalucía-Consejería de Economía y Conocimiento [Proyecto B-SEJ-135-UGR18] and the Spanish Ministry of Science and Innovation [PID2019-104239GB-I00/SRA - State Research Agency/10.13039/501100011033].