The primary goal of the present study was to assess the psychometric properties of the Arabic translation of the DASS-Y 21 items among adolescents in the Arabic-speaking community in Lebanon. The secondary aim was to develop and validate a new shortened version of the DASS-Y composed of only 12 items. Our findings indicated that both versions exhibited a three-factor structure, demonstrated good reliability, showed convergent validity, and exhibited gender invariance. These results suggest that these instruments can effectively measure psychological distress, encompassing stress, anxiety, and depression, in Arabic-speaking adolescents, particularly within the Lebanese context. Lebanon has recently experienced unprecedented social, political, and economic crises, along with a resurgence of violent extremism and a series of disasters such as the Beirut blast [45–47]. Given these unique circumstances, utilizing the Arabic DASS-Y adapted to the Lebanese context in both a full-length and a brief version may aid in identifying adolescents with elevated psychological distress levels who could benefit from targeted interventions.
Results revealed that both EFA and CFA yielded three-factor structure for the Arabic 21-item DASS-Y that mirrors the original three factors captured by the original validation study [13], and the Chinese validation [14]. Moreover, the Arabic 21-item DASS-Y showed a good internal consistency as evidenced through the McDonald’s ω values ranging from 0.92 to 0.94 and Cronbach α values ranging from .86 to .92 for the three subscales. These results are comparable to the original validation study [13] where the internal consistency was attested by Cronbach α values ranging from 0.84 to 0.89, and by McDonald’s ω values ranging from 0.84 to 0.90. After shortening the scale, we found good model fit for the three-factor solution, and adequate composite reliability, with values ranging from 0.86 to 0.92, attested by McDonald’s ω. Hence, one potential strength of the DASS-Y scale is that it assesses the three universally consensual dimensions of the psychological distress (depression, anxiety, stress) through only 12 items, while maintaining good psychometric properties.
Our results indicate evidence for measurement invariance across gender, proving that the 12- and 21-item DASS-Y can be applied to make valid comparisons between boy and girl respondents. Prior to delving into our findings, it is crucial to emphasize that according to the original study of the DASS-Y) [13], girls consistently achieved higher scores than boys across all three DASS-Y scales during high school. However, there was no discernible disparity between genders in primary school. This trend aligns closely with prior research indicating that gender discrepancies in negative affect and associated disorders typically surface during adolescence [48, 49]. In our study, for the long and short forms, higher stress scores were found in girls compared to boys, in line with previous studies [50, 51]. Actually, girls appear to face a higher susceptibility to psychological issues, stemming from a blend of biological and social factors; these factors encompass gender stereotypes, inequity, social segregation, and a lack of autonomy [52]. As for depression, no significant difference was found between girls and boys in both forms. The findings from various studies indicate gender disparities in depression, with girls exhibiting a significantly higher likelihood of experiencing depression compared to boys [53, 54]. However, it is essential to note that gender differences are not consistent across all race-ethnic groups [55]. Subsequently, females tend to predominate in the incidence of depressive disorders, but there seems to be no gender difference in terms of recurrence, remission, or chronicity of depression [56]. Finally, higher anxiety was found in girls compared to boys only in the short form, in line with the findings reported by Ohannessian et al. [57]; the latter conducted a prospective study spanning 2 years and observed that in adolescent girls, generalized anxiety and social anxiety increased from middle to late adolescence, whereas boys exhibited stable patterns.
In DASS-Y long and short forms, results showed that higher suicidal ideation was significantly associated with higher depression, anxiety and stress, in line with previous studies correlating higher suicidal ideation with higher psychological distress [58, 59]. Adolescents are particularly susceptible to experiencing depressive and anxiety disorders due to dynamic changes in both their biology and emotions (such as emotional lability), as well as social factors (such as perceived lack of support and impulsive actions) [60]. These factors contribute to elevated stress levels [61, 62], making it challenging for young individuals to adapt to the evolving demands of their surroundings [60]. Consequently, a significant percentage of suicide attempts among adolescents are associated with diagnosed anxiety disorders [63–65] and depressive disorders [65–67].
In both forms, higher academic performance was significantly associated with lower depression, anxiety and stress, in line with previous studies [68–70]. First, symptoms associated with depression, such as tiredness, lack of concentration, poor memory, and reduced attention span, negatively impact the learning ability [71]. Moreover, students facing depression tend to miss more classes, tests, and assignments, are more prone to drop courses [72], and have a higher likelihood of school dropout [73]. Consequently, depressed students may find themselves in a cyclical pattern where depression disrupts academic studies, and poor academic performance contributes to a low mood [74].
These findings confirm discriminant validity of the DASS-Y 21 and 12-item; and are in line with previous literature, as mentioned. However, to further confirm the clinical utility of the Arabic DASS-Y (long and short forms), additional validation studies in clinical populations are required.
Limitations
Our findings should be considered preliminary due to several limitations. The study utilized a cross-sectional design and relied on self-reported data. It was conducted exclusively in Lebanon, thereby restricting the generalizability of our results to Arab-speaking individuals in other Arab countries. To overcome these limitations, future research should adopt longitudinal and cross-cultural approaches. Additionally, a selection bias may exist due to the snowball sampling technique employed for participant recruitment, thereby limiting the generalizability of our findings. Future research should explore additional psychometric properties, including test-retest reliability, convergent and divergent validity, as well as content and criterion validity (predictive validity).