Further Validation of the Inventory of Callous–Unemotional Traits in Chinese Children: Cross-Informants Invariance and Longitudinal Invariance

Abstract

The present study examined the factor structure and measurement invariance of the shortened versions of the Inventory of Callous–Unemotional Traits (ICU) with data from multiple informants. Five short versions of the ICU proposed in previous studies were tested and compared through confirmatory factor analysis. The measurement invariance across different informants (i.e., self-report, parent-report, and teacher-report) and longitudinal measurement invariance for the resulting best-fitting model were tested thoroughly. Results indicated that a shortened form that consists of 11 items (ICU-11) to assess callousness and uncaring factors had excellent overall fit. Moreover, the ICU-11 was invariant across informant and occasions. However, the ICU-11 was not without limitations; the internal consistency α for the uncaring factor with self-report scores was marginal. In conclusion, our findings suggest that the ICU-11 was an excellent fit for our data and displayed measurement invariance across informants and over time. The ICU-11 may be a promising assessment tool that could be used in research to assess callous–uncaring traits in Chinese children.

Assessing Callous-Unemotional Traits in Chinese Detained Boys: Factor Structure and Construct Validity of the Inventory of Callous-Unemotional Traits

“…However, a recent meta-analysis by Deng et al (2019) has noted that there remains a lack of evidence of the applicability of the ICU among non-European-American samples. Although there has been an attempt of validating the ICU among Chinese community samples (Wang et al, 2017b, 2019), little is known of the utility of the ICU in clinical settings in non-English-speaking delinquent populations.…”

“…Furthermore, although the ICU was originally developed as a unidimensional measure of CU traits (an overarching CU factor containing three subfactors: unemotional, callousness and uncaring), this early proposed three-factor, as well as a three-factor bifactor model (Essau et al, 2006), received limited support in either community (Ciucci et al, 2014; Wang et al, 2017b, 2019) or delinquent samples (e.g., Kimonis et al, 2008) due to the poor overall fit of these models. Notably, the unemotional factor has been shown to have relatively poor psychometric properties, showing low reliability, poor factor loadings and inadequate correlations with external criteria (e.g., Essau et al, 2006; Kimonis et al, 2008; Byrd et al, 2013).…”

“…The ICU-12 has two correlated factors: callousness (seven items) and uncaring (five items), and its validity and reliability were supported in a number of subsequent studies that used detained samples (e.g., Colins et al, 2016; Paiva-Salisbury et al, 2017). Two recent studies found that an 11-item model (ICU-11) which excluded the item, “I do not show my emotions to others” – the only item retained from the unemotional factor – achieved a better fit than the ICU-12 among Chinese-speaking samples using university students (Wang et al, 2017b) and community children (Wang et al, 2019). This is possibly due to the fact that expressing emotion is generally not encouraged in Chinese culture, thus resulting in the low discriminability of the item among Chinese populations.…”

“…This is possibly due to the fact that expressing emotion is generally not encouraged in Chinese culture, thus resulting in the low discriminability of the item among Chinese populations. Nevertheless, the ICU-11 displayed measurement invariance across informants and occasions and had strong evidence for its criteria validity (Wang et al, 2019). The results of Wang et al (2017b) also showed strong associations with other measures of psychopathic traits, and both of the two factors (callousness and uncaring) correlated significantly with the total scores on the ASPD and proactive aggression.…”

“…Confirmatory factor analyses (CFA) were conducted to compare various factor structures proposed in previous studies. Based on findings from recent studies (Wang et al, 2017b, 2019), we hypothesized that the ICU-11 with the callousness and uncaring dimensions would be the best fit for the data.…”