Remapping the foundations of morality: Well-fitting structural model of the Moral Foundations Questionnaire

Abstract

Moral foundations theory posits five moral foundations, however 5-factor models provide poor fit to the data. Here, in five studies, each with large samples (total N = 11,496), we construct and replicate a well-fitting model of the Moral Foundations Questionnaire (MFQ). In study 1 (N = 2,271) we tested previously theorised models, confirming none provide adequate fit. We then developed a well-fitting model of the MFQ. This required 7 domains resulted from splitting the loyalty/purity and loyalty/patriotism aspects of existing domains, adding a general moral factor as well as a moral “tilt” factor, corresponding to left-right political distinctions. The new model replicated in 4 independent samples including western and eastern countries (N = 9,225). These findings demonstrate the first well-fitting replicable model of the MFQ. They also highlight the importance of modelling measurement structure, improving measurement precision and revealing important additional factors and group-factor structure within the moral foundations questionnaire.

The shallow of your smile: the ethics of expressive vocal deep-fakes

“…We opt to conform to the recommendations of a more extensive validation (with a sample size nearly twice as big as our current sample [ 69 ]) and use the 5-factor model for our current analysis. However, the present data adds evidence to the fact that, as already discussed elsewhere [ 71 ], significant elements of the MFQ covariance structure are not captured by this model.…”

“…A previous study [ 69 ] validated MFQ-French on a sample of similar demographics as the present study and found it had acceptable internal validity ( N = 538 participants; care: Cronbach α = 0.64; fairness: α = 0.67; loyalty: α = 0.65; authority: α = 0.73; purity: α = 0.79) and one-month test-retest validity ( N = 40; care: , 95% CI [0.26,0.72]; fairness: ; loyalty: ; authority: , [0.57, 0.86], purity: ; all ). The fit to a 5-factor structure, while significantly better than an alternative 3-factor model, was comparably poorer ( N = 538; Comparative Fit Index, CFI: 0.82; root mean squared error of approximation, RMSEA: 0.065), a known issue common to the American version and discussed elsewhere [ 70 , 71 ].…”

Measuring morality: Analyzing the factor structure of the moral foundations questionnaire in veterans

“…Findings suggest that the MFQ in its current form may correspond to a specific cultural time and place, rather than general human behavior, as evidenced by difficulty replicating the five-factor model in these studies. In light of these critiques, another recent study tested existing models of the MFQ and found them to be a poor fit but was able to produce a well-fitting alternative model comprised of seven factors using the same items (Zakharin & Bates, 2021).…”

“…Consequently, the purpose of the present study was to evaluate the psychometric properties of the MFQ with a group of US veterans. Our aims were to: (1) descriptively compare MFQ factor scores for a veteran sample to the general US population using previously reported information; and (2) test whether existing MFQ models (the original five-factor structure and the alternative seven-factor structure proposed by Zakharin & Bates, 2021) would be supported for this distinct cultural group.…”