The factor structure and psychometric properties of the Chinese version of the Positive and Negative Suicide Ideation Inventory (PANSI-C) in a non-clinical sample of Chinese adolescents

Abstract

Backgrounds The Positive and Negative Suicide Ideation (PANSI) Inventory is a widely used self-report questionnaire which is designed to comprehensively evaluate the protective factors and negative risk factors associated with suicidal behaviors among adolescents. The present study aimed to evaluate the psychometric properties and measurement invariance of the Chinese version of the PANSI in a non-clinical sample of Chinese adolescents. Methods Participants (N = 1198) were Chinese middle school students aged 11–17 years (44.8% boys and 51.9% girls, 3.3% missing values) in Guizhou Province. All participants completed the Chinese version of the Positive and Negative Suicide Ideation Inventory (PANSI-C), the Rosenberg self-esteem scale (RSE), and the suicide probability scale (SPS). Cronbach’s alpha coefficients, confirmatory factor analysis, Pearson’s correlations, and multigroup confirmatory factor analysis tests were conducted thereafter. Results The results showed that the Cronbach’s alpha coefficients for the two subscales of the PANSI-positive suicide ideation and the PANSI-negative suicide ideation were .696 and .915, respectively. The confirmatory factor analysis supported the fit of the two-factor model as the best fitting model [Chi-square goodness of fit = 703.859, p < .001, degrees of freedom = 76, comparative fit index = .919, Tucker–Lewis index = .903, standardized root mean square residual = .047, root mean square error of approximation (90% CI) = .083 (.077, .089)]. Positive suicide ideation had negative correlations with the SPS and positive correlations with the RSE, whereas the negative suicide ideation had positive correlations with the SPS and negative correlations with the RSE. All correlations were statistically significant (p < .001), demonstrating the criterion validity of the PANSI-C. Moreover, the strict measurement invariance of the PANSI-C was supported across gender, single-parent and non-single-parent households groups, and the strong measurement invariance was supported across age. Limitations The feasibility of this study is limited to Chinese normal adolescents and lack of clinical samples. Conclusion Empirical support for the reliability and validity of the PANSI-C was found. The PANSI-C instrument is found to be useful in assessing positive and negative suicide ideation in Chinese normal adolescents.

The Positive and Negative Suicidal Ideation Inventory among Portuguese Adolescents: Factor Structure and Gender Invariance

“…Invariance results indicate that the PANSI has strong gender invariance in this population. Although these results are not as conclusive as the strict invariance found by Chen et al [27], these results are sufficient to make meaningful comparisons between groups, and it can be considered that they share the same operational definition [40]. There are various possible explanations for the differences between these and Chen et al's [27] results.…”

“…Although reliability scores raise some concerns regarding Item 4 slightly raising the scale's α, the literature regarding the adolescent version of the PANSI shows that its items do not factor as well into their respective scales, as is the case with Avendaño-Prieto et al [25], who found that certain items tend to aggregate into a third factor. Additionally, regarding positive ideation's lower reliability, Chen et al [27] argue that non-clinical samples might have a harder time understanding some of the items relating to this subscale. These results, in addition to positive ideation's relatively low AVE, highlight the difficulty of assessing suicidal protective factors through self-report in non-clinical populations.…”

“…In Portugal, the psychometric characteristics of the PANSI have been studied in young adults [22] but not in adolescents. The PANSI has been adapted for various cultures (e.g., Villalobos-Galvis [23]) and populations (e.g., Osman et al [24]), consistently retaining its two-factor structure spread across fourteen items, including in children and adolescents [25][26][27][28][29]. The PANSI shows particular utility due to its widespread validation, its multidimensional assessment, and its briefness (14 items), leading to ease of application.…”

“…As it stands, using the PANSI in the assessment of suicide risk is particularly advantageous due to its ability to give a multifaceted assessment by also capturing protective factors, instead of the common approach that focuses solely on assessing risk factors [27]. Another advantage the PANSI might present is related to its invariance across gender.…”

“…Recognizing and addressing group differences is essential to promoting equity in public health and research, and failing to do so can negatively affect boys and, especially, girls [33], as is evidenced by a lack of data regarding the risk of death by suicide for girls [32]. Chen et al [27] showed strict invariance across gender in a Chinese adolescent sample, demonstrating the PANSI's lack of gender bias in its factor structure. Reproducing and expanding on these results is the next step in promoting an equitable practice in this field and, as such, the present study aims to achieve the following objectives: (1) to assess the reliability and validity of the PANSI scale in a non-clinical sample of Portuguese adolescent students; (2) assess the internal structure of the PANSI through confirmatory factorial analysis; and (3) test the measurement invariance of the PANSI scale across gender.…”