Validation of the Korean Version of the Center for Epidemiologic Studies Depression Scale-Revised in Korean Adolescents

Ji-Min Kim; Kyeong-Sook Choi; Ji Ae Nam; Hyun Woo Han; Je-Chun Yu

doi:10.30773/pi.2024.0285

Psychiatry Investig > Volume 22(11); 2025 > Article

Kim, Choi, Nam, Han, and Yu: Validation of the Korean Version of the Center for Epidemiologic Studies Depression Scale-Revised in Korean Adolescents

Original Article

Psychiatry Investigation 2025;22(11):1290-1299.

Published online: November 12, 2025

DOI: https://doi.org/10.30773/pi.2024.0285

Validation of the Korean Version of the Center for Epidemiologic Studies Depression Scale-Revised in Korean Adolescents

Ji-Min Kim¹

, Kyeong-Sook Choi^1,^2,³

, Ji Ae Nam^3,⁴

, Hyun Woo Han¹

, Je-Chun Yu^1,²

¹Department of Neuropsychiatry, Eulji University Hospital, Daejeon, Republic of Korea

²Department of Neuropsychiatry, Eulji University School of Medicine, Daejeon, Republic of Korea

³Cheongdang Clear Mind Psychiatric Clinic, Cheonan, Republic of Korea

⁴Daedeok-gu Mental Health Welfare Center, Daejeon, Republic of Korea

Correspondence: Kyeong-Sook Choi, MD, PhD Cheongdang Clear Mind Psychiatric Clinic, Master Building 702, 18 Cheongsu 11-ro, Dongnam-gu, Cheonan 31198, Republic of Korea Tel: +82-41-414-0707, Fax: +82-41-414-0708 E-mail: cksinj@gmail.com

Correspondence: Je-Chun Yu, MD Department of Neuropsychiatry, Eulji University Hospital, 95 Dunsanseo-ro, Seo-gu, Daejeon 35233, Republic of Korea Tel: +82-42-611-3443, Fax: +82-42-611-3445, E-mail: jcyu@eulji.ac.kr

Received September 5, 2024 Revised January 31, 2025 Accepted February 6, 2025

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Objective

The Korean version of the Center for Epidemiologic Studies Depression Scale-Revised (K-CESD-R) is widely used in community epidemiology studies to screen depressive symptoms, given its proven reliability in diverse populations. However, it has not yet been validated in Korean adolescents. Therefore, this study validated and standardized the K-CESD-R for use in Korean adolescents.

Methods

Data were collected from 2,419 adolescents aged 12-17 years in educational institutions across Daedeok District, Daejeon, South Korea. To evaluate reliability, the internal consistency of the K-CESD-R was measured using Cronbach’s alpha. Concurrent validity was tested through Pearson correlation analysis of established scales, and construct validity was assessed via exploratory factor analysis (EFA) and confirmatory factor analysis (CFA).

Results

The K-CESD-R had high internal consistency (Cronbach’s alpha=0.934). Correlation analyses supported strong concurrent validity with other depression scales, although there was a negative correlation with the self-esteem scale. Factor analysis revealed a three-factor structure that explained 61.792% of the total variance. CFA of the K-CESD-R using a structural equation model affirmed the three-factor structure hypothesized in our EFA. The model fit indices demonstrated acceptable levels of fit, including Root Mean Square Error of Approximation=0.093, Comparative Fit Index=0.886, and Tucker-Lewis Index=0.871.

Conclusion

The K-CESD-R is a valid and reliable instrument for screening for depressive symptoms in Korean adolescents, supporting its use in epidemiological research and clinical settings. Further research should explore its applicability across Korean adolescent populations to confirm these findings.

Keywords: Adolescent psychiatry, Depression, Health survey, Psychometrics

INTRODUCTION

Depression is among the most common mental disorders affecting adolescents worldwide [1]. According to the World Health Organization, depression was diagnosed in 3.9% of adolescents globally in 2021 [2]. By contrast, a study conducted in Seoul in 2005 found that the self-reported depression rate among adolescents was 7.37% [3]. A meta-analysis conducted from 2001 to 2020 found that the Middle East, Africa, and Asia had the highest rates of depressive symptoms among adolescents [1]. More recently, in 2022, a South Korean study found that approximately 28.7% of middle and high school students reported experiencing depressive symptoms within the past year [4].

Adolescents experience depressive symptoms differently due to their unique developmental stage. Adolescence is characterized by rapid biological and psychosocial changes, including hormonal fluctuations, brain maturation, and evolving identity formation, which can amplify vulnerability to depression [5]. Hormonal changes, such as increased cortisol and sex hormones, can affect emotional regulation, heightening the risk of depressive symptoms [6]. Additionally, adolescents may exhibit distinct depressive symptoms, such as irritability and aggression, compared to adults [7]. Even subclinical depressive symptoms during adolescence can be a developmental challenge, leading to enduring consequences for social functioning, academic performance, and future employability [8,9].

Therefore, timely intervention and treatment depend heavily on the prompt identification and precise evaluation of depressive symptoms [10,11]. Early intervention in school-based programs can prevent adolescents from developing risk factors for unemployment, suicidal thoughts, substance abuse, and depression as adults [10]. Screening for mental disorders in this demographic is crucial given the significant costs associated with healthcare, education, mental health services, and the judicial system [12]. The need for targeted screening tools tailored to the developmental characteristics of adolescents is critical, as it enables early detection and intervention, mitigating the long-term impact of untreated depressive symptoms.

The Center for Epidemiologic Studies Depression Scale (CES-D) has been widely used across diverse adult populations due to its effectiveness in depression screening [13,14]. However, with changing psychiatric diagnostic criteria, eight CES-D items have become outdated with respect to the current definition of major depression [15]. Therefore, the Center for Epidemiologic Studies Depression Scale-Revised (CESD-R) [8], a revised version of the CES-D, was developed to cover the depression criteria in the Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV-TR and DSM-5 classifications [11,15]. The CESD-R has the advantage of being freely accessible without copyright restrictions and has been translated into various languages [16].

The Korean version of the Center for Epidemiologic Studies Depression Scale-Revised (K-CESD-R) is extensively used in community epidemiology studies and clinical settings across Korea [16]. This widespread use of the K-CESD-R is due to its proven internal consistency, retest reliability, and validity in screening and monitoring depressive symptoms in adult populations [17]. However, its use in Korean adolescents has not been validated. Therefore, this study sought to validate and standardize the K-CESD-R using a large sample of Korean adolescents.

We tested the effectiveness of the scale in a school-based mental health program, where it was used to screen adolescents for depressive symptoms. By conducting this study, we seek to confirm the robustness of this tool for the early detection of depressive symptoms, thereby contributing to better mental health outcomes for adolescents.

METHODS

Data source and participants

Data were obtained from adolescents aged 12-17 years who participated in a school-based mental health intervention program implemented by Daedeok District Mental Health Welfare Center in Daejeon, South Korea, from 2017 to 2023. This program is implemented every year as studies show that adolescents enrolled in the Korean academic system have better mental health [10]. The program enrolls students who agree to participate from schools that voluntarily apply to be part of the initiative. This research was approved by the Eulji University Hospital Institutional Review Board (EMC 2024-06-008).

Measures

CES-D

CES-D is a 20-item self-report questionnaire originally developed by Radloff [13] in 1977 to measure depressive symptoms in the general population. Participants rate the frequency of symptoms experienced over the past week on a 4-point Likert scale ranging from 0 (rarely or none of the time) to 3 (most or all of the time). This scale has been widely validated across diverse populations for its reliability and utility in epidemiological studies.

The Korean-translated version of the CES-D, adapted by Cho and Kim [18] has demonstrated strong internal consistency, with a Cronbach’s α of 0.90. The CES-D has also been validated in Korean adolescents by Heo et al. [19], demonstrating high internal consistency (Cronbach’s α=0.88) and construct validity. The CES-D was used to assess the concurrent validity of the K-CESD-R, providing a validated comparative measure for depressive symptoms.

K-CESD-R

In 1977, Radloff [13] developed the CES-D to assess depressive symptoms. In 2004, Eaton et al. 15 revised the scale: the CESD-R incorporates the nine principal symptoms of a major depressive episode, as defined by the DSM-IV, persisting for >2 weeks. CESD-R scores are strongly correlated with scores on the original CES-D, showing that it is a reliable tool for assessing depressive symptoms across various international populations [17]. The K-CESD-R, standardized by Lee et al. [16] in 2016 for the general population, was adapted to the specific epidemiological and cultural context of Korea. The K-CESD-R is a concise self-administered questionnaire consisting of 20 items, each rated on a 5-point Likert scale, resulting in a total score ranging from 0 to 80. In adults, the scale had a Cronbach’s alpha of 0.98, indicating high reliability, and it was found to have a two-factor structure.

Korean Social Anxiety Scale for Children and Adolescents

The Korean Social Anxiety Scale for Children and Adolescents (K-SAS-A) was developed by Moon and Oh [20] as a tool for measuring social anxiety in Korean children and adolescents. The scale was constructed based on the Social Phobia and Anxiety Inventory for Children [21] and the Social Anxiety Scale for Children-Revised [22]. The K-SAS-A is comprised of 40 items evaluated on a 5-point Likert scale, with a total score that ranges from 0 to 200 [23]. In the original study, Cronbach’s alpha ranged from 0.79 to 0.92, demonstrating strong internal consistency [20].

State Anxiety Inventory for Children

Spielberger originally developed the State Anxiety Inventory for Children (SAIC) to measure situational and enduring anxiety levels in children [24]. The SAIC focuses on state anxiety, which is defined as the experience of anxiety as a response to specific situations or stimuli; in contrast, trait anxiety is a more stable aspect of personality?. Cho and Choi [25] translated and standardized the SAIC for use with Korean elementary school students, taking account of cultural relevance and allowing application in educational and clinical settings in Korea. The SAIC consists of 20 items, each evaluated using a 3-point Likert scale, yielding a total score of 0-60. Higher scores correspond to greater levels of state anxiety. The median Cronbach’s alpha for the state anxiety subscale (S-Anxiety) was 0.93, while the trait anxiety subscale (T-Anxiety) had a median Cronbach’s alpha of 0.90, both demonstrating strong internal consistency [26].

Reynolds’ Suicidal Ideation Questionnaire

Reynolds developed the Reynolds’ Suicidal Ideation Questionnaire (SIQ) to screen for suicidal ideation in adolescents and troubled youths by assessing the frequency of suicidal thoughts [27]. The SIQ is an essential tool for evaluating adolescent mental health, as suicidal ideation is more likely to occur in vulnerable populations, such as females and adolescents with depressive symptoms or conduct problems [28]. Identifying suicidal ideation at an early stage allows for timely interventions that can prevent further escalation into suicidal behaviors. In this study, we used the Korean version of the tool developed by Shin et al. [29] This self-administered questionnaire consists of 30 items, each scored on a scale ranging from 0 to 6, yielding total scores of 0-180 [30]. The scores are interpreted as follows: 62-76 indicates a higher frequency of suicidal ideation compared to peers; 77-90 suggests a significant level of suicidal ideation; and scores above 91 denote a very high level of suicidal ideation [31]. In Heo et al.’s study [19], the Cronbach’s alpha was 0.97.

Child Report of Post-traumatic Symptoms

The Child Report of Post-traumatic Symptoms (CROPS) is a self-report measure developed by Greenwald [32] to screen for post-traumatic symptoms in children and adolescents. The CROPS was created to fill a gap by evaluating post-traumatic stress symptoms beyond the strict criteria of post-traumatic stress disorder (PTSD), allowing a comprehensive understanding of how trauma affects young individuals. Each item, rated on a 3-point scale, reflects potential symptoms experienced by young individuals after trauma; the cutoff point for possible PTSD is 19. This study used the adaptation of Lee et al. [33], which had high reliability (Cronbach’s alpha of 0.91).

Korean version of the Rosenberg Self-Esteem Scale

The Rosenberg Self-Esteem Scale (RSES) is widely used for measuring self-esteem [34]. It consists of 10 items evaluated on a 4-point Likert scale, yielding scores between 10 and 40. Five items assess positive self-esteem and five assess negative selfesteem, with higher scores reflecting greater self-esteem. In this study, we used the version translated and validated by Bae et al. [35], which had a high level of reliability (Cronbach’s alpha of 0.90).

Statistical analysis

To evaluate reliability, the internal consistency of the K-CESD-R is measured using Cronbach’s alpha. Furthermore, the corrected item-total correlations were analyzed to determine the coherence between each item and the overall scale.

Validity was assessed based on both concurrent and construct validity. To assess concurrent validity, Pearson correlation was performed of established scales (CES-D, K-SAS-A, SIQ, CROPS, and RSES) to determine if the K-CESD-R scores are positively associated with those on scales assessing negative psychological constructs and inversely associated with those on scales evaluating positive psychological constructs. We used exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) to explore the factorial structure of the K-CESD-R. Principal component analysis (PCA) with Varimax rotation was used to identify the underlying factor structure of depressive symptoms. This analysis validated the primary factors identified using metrics such as the Comparative Fit Index (CFI), Tucker-Lewis index (TLI), standardized root mean square residual (SRMR), and Root Mean Square Error of Approximation (RMSEA). A p-value<0.05 was considered to indicate statistical significance.

All data analyses were done using the Statistical Package for the Social Sciences 27.0 (IBM Corp.) and R and R studio (R studio 2023.12.1 Build 402; R Foundation for Statistical Computing).

RESULTS

Sociodemographic characteristics of the study subjects

Of the original 3,053 participants, those with missing data were excluded, leaving 2,419 complete surveys for the analysis. This study utilized a complete case analysis approach, where participants with incomplete responses were excluded from the dataset. We used G^*Power 3.1.9.7 (Kiel University) to calculate the required sample size for performing CFA with a significance level of 0.05, effect size of 0.25, and power of 0.95, considering a total of four predictors. The calculations suggested that a minimum sample size of 73 participants was required. Our actual sample size significantly exceeds this, ensuring statistical robustness and power and thus enhancing the reliability of our findings [36,37].

The participants included 1,268 females (52.4%) and 1,151 males (47.6%). The largest group of participants was aged 17 years (45.6%), followed by those aged 14 years (31.8%) and those aged 15 years (12.6%). Table 1 summarizes the subjects’ demographic and clinical characteristics, including age, school record, parental education level, socioeconomic status, suicidal ideation, and attempt.

Reliability

Internal consistency

Cronbach’s alpha for the K-CESD-R for the entire adolescent sample was 0.934, demonstrating high reliability for assessing depressive symptoms within this population. The correlations between the K-CES-D-R items for Korean adolescents are detailed in Table 2. In the analysis of corrected item-total correlations, most of the 20 items had strong relationships with the total score, with values exceeding 0.40, indicating good alignment with the overall scale. However, item 18, which pertains to losing weight, had a lower correlation with the total score (corrected item-total correlations=0.302), i.e., a weaker association compared to the other items.

Validity

The validity of the K-CESD-R was evaluated through analyses of concurrent and construct validity. Concurrent validity was assessed via Pearson correlation analysis of the relationship between the K-CESD-R and several established scales that have been standardized and validated in Korean, including the CES-D, K-SAS-A, SAIC, SIQ, CROPS, and K-RSES. The correlations between the K-CESD-R and these scales were all significant (p<0.01). The K-CESD-R was positively correlated with the first five scales measuring negative psychological constructs. However, it was negatively correlated with the K-RSES (r=-0.589), a scale designed to measure self-esteem and positive psychological constructs (Table 3).

The construct validity of the K-CESD-R was evaluated using EFA and CFA, focusing on identification and confirmation of the underlying factor structure of the scale.

EFA

Initially, the factor structure of the K-CESD-R was explored using PCA with Varimax rotation. The EFA revealed a three-factor structure. Each factor with an eigenvalue >1 was retained, as these factors contribute meaningfully to the variance in the dataset, corresponding to distinct dimensions of depression. Collectively, Factors I-III explained 61.792% of the total variance, accounted for 48.096%, 7.850%, and 5.846% of the variance, respectively (Table 4). Factor I primarily consisted of items related to negative affect, such as “I felt depressed,” “I felt sad,” and “I did not like myself.” Factor II included items related to cognitive difficulties, such as trouble concentrating and feeling slowed down, as well as general fatigue. Factor III comprised somatic symptoms like sleep disturbances and appetite changes.

CFA

Following the EFA, CFA was conducted to validate the three-factor structure, using a structural equation model to assess the fit of the proposed factor structure. The analysis yielded satisfactory results, indicating that the three-factor model adequately fits the data. The factor loadings for each item were substantial, suggesting strong associations between the observed variables and their corresponding latent constructs. The loadings are represented visually in Figure 1, where Factors I-III were highly correlated with their respective items.

The CFA results were assessed using the CFI, TLI, SRMR, and RMSEA fit indices, which supported the adequacy of the three-factor model in explaining the underlying structure of the K-CESD-R within the study population, demonstrating levels comparable to those found in similar scales used in different cultural and demographic contexts (Table 5).

The inter-factor correlations between the factors were strong, particularly between Factors I and II (r=0.85), as indicated in Figure 1.

DISCUSSION

This study sought to validate the K-CESD-R in a large sample of Korean adolescents. The study found strong evidence supporting the reliability and validity of the K-CESD-R as a tool for assessing depressive symptoms in adolescents.

The K-CESD-R had high internal consistency, with a Cronbach’s alpha of 0.934, indicating that the scale is reliable for use in adolescents. The analysis of corrected item-total correlations further supported the reliability of the scale, with most items showing strong correlations with the total score. While prior studies in Korean adults reported that item 11 (hypersomnia) had a lower corrected item-total correlation, our study found that item 18 (weight loss) had the weakest correlation in adolescents. This finding may reflect developmental differences; adolescents’ weight changes are influenced by nondepressive factors such as dieting, sports activities, or growth spurts, which can obscure the relationship between weight loss and depression. While dieting and sports activities have been noted as influencing weight changes, the biological and psychosocial characteristics of adolescence provide additional layers of complexity. Developmental changes during adolescence, such as hormonal fluctuations and pubertal growth, can also obscure the relationship between weight loss and depression, as these biological processes naturally lead to significant weight variability with gender-specific differences in fat and muscle development. These biological changes often interact with psychological factors like body dissatisfaction, a prevalent issue among adolescents, particularly females. These factors might dilute the association between weight loss and depression, making it a less reliable symptom in this population compared to adults, where unintentional weight loss may more directly signal depressive symptoms. Cultural factors could also play a role. According to Lee et al. [38], Korean adolescents commonly perceived their body shape as being “normal” or “slightly overweight,” while objective body mass index measurements revealed that a significant proportion of these adolescents were actually classified as “underweight.” This discrepancy suggests that there is considerable distortion in how Korean adolescents perceive their body shapes compared to their actual physical condition. In a society where there is considerable focus on physical appearance, weight loss may be perceived positively by adolescents, potentially leading to underreporting or misinterpretation thereof in the context of depression. This highlights the need for biological, psychological, and behavioral appropriate adaptations of depressive symptomatology assessments in different age groups.

The concurrent validity of the K-CESD-R was confirmed herein by its significant correlations with other established psychological scales. The positive correlations with scales measuring negative psychological constructs, such as the CES-D, K-SAS-A, SAIC, SIQ, and CROPS, indicate that the K-CESD-R effectively captures depressive symptoms, as it aligns closely with other tools measuring similar constructs. The K-CESD-R had strong correlations with the original CES-D, and with scales assessing suicidal ideation and trauma. The strong correlation with the CES-D further underscores the scale’s validity in assessing depressive symptoms consistent with established measures. These findings also align with previous research conducted in the United Kingdom, which reported that individuals who have experienced abuse are more likely to report severe depression, anxiety, and suicidal behavior [39]. The moderate correlation observed between the K-CESD-R and K-SAS-A, SAIC is consistent with the literature. While both depression and anxiety involve overlapping affective components, they can be distinguished by different psychological characteristics and physical responses. Depression often involves persistent negative affect and cognitive dysfunction, while anxiety is characterized by psychological hyperarousal and fear responses [40,41]. The K-CESD-R was designed to measure depressive symptoms across multiple dimensions, including negative affect, cognitive impairments, and somatic symptoms such as appetite loss and sleep disturbances. However, the scale does not contain items that directly assess physiological arousal or fear responses, which are core components of anxiety. This limitation may explain the moderate correlation observed between the K-CESD-R and anxiety measures. The K-SAS-A primarily evaluates psychological traits related to social anxiety and fear of negative evaluation. These traits are more closely associated with specific subtypes of anxiety and may have a weaker connection to depressive features, such as anhedonia or persistent low mood. The SAIC, on the other hand, measures state anxiety, emphasizing transient anxiety responses to specific situations. This construct differs from the more chronic and pervasive nature of depressive symptoms, which include sustained negative affect and emotional blunting. The episodic nature of state anxiety makes it less aligned with the long-term characteristics of depression. This underscores the importance of using distinct yet complementary tools to evaluate depression and anxiety effectively within this demographic. Conversely, the negative correlation with the RSES shows the expected inverse relationship between depression and self-esteem. These results reinforce the utility of the K-CESD-R for differentiating between positive and negative psychological constructs, supporting its use as a reliable tool for evaluating depressive symptoms.

The construct validity of the K-CESD-R was explored using both EFA and CFA. EFA revealed a three-factor structure that explained 61.792% of the total variance. This structure aligns with the theoretical underpinnings of depressive symptoms, with distinct factors representing different dimensions of depression. Some differences were observed when compared to studies conducted on different populations. A Korean study that performed a factor analysis of the general Korean population identified a two-factor model [16]. In contrast, the original English scale [42] and a Japanese study [17] both reported unidimensional factor structures. In our study, Factor I (negative mood) includes items such as “I wished I were dead,” “I wanted to hurt myself,” and “I did not like myself,” and is similar to the negative mood cluster identified in a study of the general US population using the CESD-R [42]. The functional impairment cluster in the US study, which broadly captured the impact of depressive symptoms on daily functioning, was separated into Factor II (cognitive and psychomotor impairment) and Factor III (sleep disturbances and appetite changes) in our analysis. Adolescence is a critical developmental period characterized by rapid brain maturation, particularly in the prefrontal cortex and limbic system [43,44]. The incomplete integration of these regions may lead adolescents to perceive cognitive and somatic symptoms as related but separate domains due to the differential processing of depressive dimensions within neural networks [45,46].

These discrepancies also reflect the unique developmental, biological, and methodological contexts of Korean adolescents. Factor II includes items related to trouble concentrating, feeling slowed down, and general fatigue. Academic stress and social comparison, prevalent in Korean adolescents, appear to influence the manifestation of depressive symptoms uniquely [47,48]. Cognitive impairments, such as difficulty concentrating and mental fatigue, are likely intensified by intense academic pressures, rendering these symptoms prominent enough to emerge as an independent factor. Similarly, somatic symptoms, such as sleep disturbances and appetite changes, often arise in response to stress but also impacted by hormonal changes and growth-related processes during adolescence as previously described, which may have contributed to its separation as an independent factor. Previous K-CESD-R study has focused on clinical or general adult populations, where depressive symptoms may present as less differentiated, leading to two-factor solutions [16]. In contrast, our study targeted a nonclinical adolescent population, where the intensity and differentiation of symptoms may vary due to developmental and environmental factors. The sample composition also has contributed to the identification of cognitive and somatic impairments as distinct factors.

The CFA provided additional support for the three-factor model identified by the EFA. The fit indices from the CFA, showing that the RMSEA was slightly above 0.08 while both the CFI and TLI fell just below the 0.90 threshold, still indicate an acceptable fit of the model to the data, confirming that the three-dimensional structure of the K-CESD-R is appropriate for assessing depressive symptoms in Korean adolescents (RMSEA=0.093, CFI=0.886, TLI=0.871). The fit index was slightly lower but similar to that of adolescents in other countries [42]. The inter-factor correlations between the factors were strong, particularly between Factors I and II (r=0.85), indicating that while distinct, these factors are closely related within the construct of depressive symptoms (Figure 1). This close relationship may be attributed to the intertwined nature of emotional and cognitive processes during adolescence. Research indicates that fluctuation in mood states can affect cognitive functions such as attention and memory, leading to impairments in learning transfer and decision-making [49,50]. Conversely, cognitive distortions and difficulties can exacerbate negative perceived stress, creating a reciprocal cycle that intensifies both mood and cognitive symptoms [51]. Additionally, adolescent depressive patients show elevated amygdala activity to negative mood and decreased hippocampal volumes, which are critical regions involved in emotion regulation and cognitive control [43,44]. The incomplete integration of these neural systems during this period may result in a heightened interdependence of mood and cognitive symptoms, as adolescents may struggle with effectively managing emotional distress, leading to cognitive disruptions, and vice versa [52].

In conclusion, our validation of the K-CESD-R for use among Korean adolescents has implications for both clinical practice and research. Clinically, the K-CESD-R can be used as a reliable and valid tool for the early detection and monitoring of depressive symptoms in Korean adolescents, facilitating timely interventions. Early identification can enable timely interventions, potentially reducing the risk of adverse outcomes such as academic failure, substance use, and suicidal behavior. However, given the cultural and developmental nuances identified in this study, clinicians should exercise caution when interpreting certain symptoms, such as weight loss, and consider the broader psychosocial context.

This study has several limitations. The sample was recruited from a single geographic region. This recruitment strategy likely introduced regional biases, limiting the generalizability of the findings to all Korean adolescents. Differences in socioeconomic status, cultural practices, and educational environments across regions may influence how depressive symptoms are expressed and reported. Cultural perceptions of mental health and stigma associated with depression could also have influenced self-reported responses. Thus, caution should be exercised when applying these findings to a broader population. Limited clinical characteristics are included due to the focus of this study on validating the K-CESD-R in a non-clinical sample; future research should explore associations between demographic and clinical variables to provide a more comprehensive understanding of the population. Missing data was excluded by Complete Case Analysis. This method ensures robust statistical analysis by relying solely on fully completed data. However, it may introduce selection bias if the missing data are not completely random, such as being related to the student’s low level of participation or engagement. While this approach is commonly used in validation studies, it reduces the overall sample size and may limit the generalizability of findings to the broader population. This can be addressed by employing advanced imputation techniques to account for missing data.

Future research should further explore the applicability of the K-CESD-R across different subgroups of adolescents, including those with various socioeconomic backgrounds residing in different regions of Korea. This approach could enhance the accuracy and reliability of depression screening tools, ensuring they are sensitive to the unique experiences of each demographic. Additionally, longitudinal studies that examine the stability of the factor structure and the predictive utility of the K-CESD-R over time are necessary to provide further evidence of its validity and reliability in assessing adolescent depression.

Notes

Availability of Data and Material

The datasets generated or analyzed during the study are not publicly available due to privacy concerms, but are available from the corresponding author on reasonable request.

Conflicts of Interest

The authors have no potential conflicts of interest to disclose.

Author Contributions

Conceptualization: Kyeong-Sook Choi. Data curation: Ji Ae Nam, Kyeong-Sook Choi. Formal analysis: Ji-Min Kim, Kyeong-Sook Choi, Hyun Woo Han. Investigation: Ji Ae Nam, Kyeong-Sook Choi. Methodology: Kyeong-Sook Choi, Ji-Min Kim. Project administration: Kyeong-Sook Choi. Software: Kyeong-Sook Choi. Supervision: Kyeong-Sook Choi, Je- Chun Yu. Validation: Kyeong-Sook Choi, Ji-Min Kim. Visualization: Ji-Min Kim, Kyeong-Sook Choi. Writing—original draft: Ji-Min Kim, Kyeong-Sook Choi. Writing—review & editing: Kyeong-Sook Choi.

Funding Statement

None

Acknowledgments

None

Figure 1.

CFA of the three-factor model for K-CESD-R. This figure represents the CFA results for the three-factor model of the KCESD- R. Factors I, II, and III correspond to distinct dimensions of depressive symptoms identified through exploratory factor analysis. The boxes on the right side (CES1-CES20) represent the 20 items from the K-CESD-R questionnaire, with each item corresponding to a specific question described in Table 2. The values adjacent to the arrows indicate the factor loadings, reflecting the strength of the relationship between each item and its respective factor. For instance, Factor I, primarily reflecting negative affect, includes items such as CES14 (“I wished I were dead”) and CES15 (“I wanted to hurt myself”), which show high loadings of 0.76 and 0.66, respectively. Inter-factor correlations are shown with curved double-headed arrows, providing insights into the relationships between the three factors. Factor I and Factor II exhibit the strongest correlation (r=0.85), suggesting a close relationship between negative mood and cognitive impairments. CFA, confirmatory factor analysis; K-CESD-R, Korean version of the Center for Epidemiologic Studies Depression Scale-Revised.

Table 1.

Demographic and clinical characteristics of subjects

Demographics	Number (%)
Sex (N=2,419)
Male	1,151 (47.6)
Female	1,268 (52.4)
Age (N=2,419)
12 years	80 (3.3)
13 years	81 (3.3)
14 years	770 (31.8)
15 years	305 (12.6)
16 years	80 (3.3)
17 years	1,103 (45.6)
School record (N=2,373)
Good	745 (30.8)
Fair	971 (40.1)
Poor	657 (27.2)
Parental education level (father) (N=2,037)
≥University	1,290 (53.3)
High school	705 (29.1)
≤Middle school	42 (1.7)
Parental education level (mother) (N=2,039)
≥University	1,260 (52.1)
High school	727 (30.1)
≤Middle school	52 (2.2)
Socioeconomic status (N=2,351)
High	790 (32.6)
Fair	1,353 (55.9)
Low	208 (8.6)
Relationship with parents (N=2,409)
Satisfied	2,058 (85.0)
Neutral	312 (12.9)
Dissatisfied	39 (1.6)
Suicidal ideation (N=2,406)
Yes	361 (14.9)
No	2,045 (84.5)
Suicidal attempt (N=2,374)
Yes	46 (1.9)
No	2,328 (96.2)

Table 2.

Corrected item-total correlations for K-CESD-R item

Item		M (SD)	Item-total correlation	Alpha if item deleted
1	My appetite was poor.	0.36 (0.76)	0.448	0.934
2	I could not shake off the blues.	0.34 (0.78)	0.770	0.928
3	I had trouble keeping my mind on what I was doing.	0.50 (0.89)	0.708	0.929
4	I felt depressed.	0.30 (0.74)	0.782	0.928
5	My sleep was restless.	0.32 (0.78)	0.519	0.933
6	I felt sad.	0.36 (0.78)	0.751	0.929
7	I could not get going.	0.32 (0.76)	0.753	0.929
8	Nothing made me happy.	0.19 (0.63)	0.681	0.930
9	I felt like a bad person.	0.19 (0.61)	0.682	0.930
10	I lost interest in my usual activities.	0.34 (0.77)	0.734	0.929
11	I slept much more than usual.	0.50 (0.89)	0.435	0.935
12	I felt like I was moving too slowly.	0.50 (0.91)	0.659	0.930
13	I felt fidgety.	0.19 (0.61)	0.687	0.930
14	I wished I were dead.	0.12 (0.53)	0.664	0.931
15	I wanted to hurt myself.	0.08 (0.46)	0.580	0.932
16	I was tired all the time.	0.96 (1.27)	0.605	0.934
17	I did not like myself.	0.26 (0.73)	0.750	0.929
18	I lost a lot of weight without trying to.	0.18 (0.61)	0.302	0.936
19	I had a lot of trouble getting to sleep.	0.27 (0.76)	0.552	0.932
20	I could not focus on the important things.	0.44 (0.86)	0.709	0.929
Internal consistency				Cronbach’s alpha=0.934

K-CESD-R, The Korean version of Center for Epidemiologic Studies Depression Scale-Revised; M, mean; SD, standard deviation.

Table 3.

Correlation of total scores of each scale with the total score of the K-CESD-R

Scales	K-CESD-R	CES-D	K-SAS-A	SAIC	RSIQ	CROPS	K-RSES
K-CESD-R	1
CES-D	0.813^**	1
K-SAS-A	0.487^**	0.565^**	1
SAIC	0.567^**	0.666^**	0.508^**	1
RSIQ	0.729^**	0.684^**	0.446^**	0.453^**	1
CROPS	0.763^**	0.747^**	0.610^**	0.587^**	0.625^**	1
K-RSES	-0.589^**	-0.728^**	-0.549^**	-0.638^**	-0.558^**	-0.611^**	1

^** p<0.01.

K-CESD-R, Korean version of Center for Epidemiologic Studies Depression Scale-Revised; CES-D, Center for Epidemiological Studies Depression Scale; K-SAS-A, Korean Social Anxiety Scale for Children and Adolescents; SAIC, State Anxiety Inventory for Children; RSIQ, Reynolds Suicide Ideation Questionnaire; CROPS, Child Report of Post-traumatic Symptoms; K-RSES, Korean version of the Rosenberg Self-Esteem Scale.

Table 4.

Factor analysis of K-CESD-R items for entire sample

Item		Factor I	Factor II	Factor III
14	I wished I were dead.	0.848	0.136	0.149
15	I wanted to hurt myself.	0.799	0.051	0.150
17	I did not like myself.	0.762	0.349	0.168
4	I felt depressed.	0.718	0.406	0.220
9	I felt like a bad person.	0.708	0.293	0.178
6	I felt sad.	0.690	0.409	0.193
8	Nothing made me happy.	0.645	0.343	0.201
2	I could not shake off the blues.	0.644	0.488	0.183
13	I felt fidgety.	0.596	0.383	0.236
12	I felt like I was moving too slowly.	0.288	0.709	0.152
20	I could not focus on the important things.	0.302	0.692	0.285
3	I had trouble keeping my mind on what I was doing.	0.311	0.690	0.269
16	I was tired all the time.	0.161	0.686	0.277
10	I lost interest in my usual activities.	0.468	0.607	0.199
11	I slept much more than usual.	0.151	0.602	-0.003
7	I could not get going.	0.552	0.575	0.168
5	My sleep was restless.	0.109	0.262	0.810
19	I had a lot of trouble getting to sleep.	0.173	0.253	0.799
1	My appetite was poor.	0.221	0.178	0.562
18	I lost a lot of weight without trying to.	0.152	0.008	0.561
Eigen-value		9.619	1.570	1.169
Variance explained (%)		48.096	7.850	5.846

K-CESD-R, Korean version of Center for Epidemiologic Studies Depression Scale-Revised.

Table 5.

Fit indices for confirmatory factor analysis

Model	χ²	df	CFI	TLI	SRMR	RMSEA
	3,695.179	167	0.886	0.871	0.059	0.093

CFI, Comparative Fit Index; TLI, Tucker-Lewis Index; SRMR, Standardized Root Mean Square Residual; RMSEA, Root Mean Square Error of Approximation.

REFERENCES

1. Shorey S, Ng ED, Wong CHJ. Global prevalence of depression and elevated depressive symptoms among adolescents: a systematic review and meta-analysis. Br J Clin Psychol 2022;61:287-305.

2. World Health Organization. Adolescent mental health. Geneva; 2021.

3. Park HY, Heo J, Subramanian SV, Kawachi I, Oh J. Socioeconomic inequalities in adolescent depression in South Korea: a multilevel analysis. PLoS One 2012;7:e47025

4. Korea Disease Control and Prevention Agency. Trends in the prevalence of perceived depressive symptoms among Korean adolescents, during 2012-2022. Publ Health Wkly Rep 2023;16:1631-1632.

5. Romeo RD. The teenage brain: the stress response and the adolescent brain. Curr Dir Psychol Sci 2013;22:140-145.

6. Naninck EF, Lucassen PJ, Bakker J. Sex differences in adolescent depression: do sex hormones determine vulnerability? J Neuroendocrinol 2011;23:383-392.

7. Ginicola MM. Children’s unique experience of depression: using a developmental approach to predict variation in symptomatology. Child Adolesc Psychiatry Ment Health 2007;1:9

8. Bahn G, Kim B, Hwang J. [A trial application of evidence-based and school-based community program for enhancement of mental health in school children]. Seoul: National Center for Mental Health; 2010,Korean.

9. Kwon MJ, Kim JE. [The mediating role of grit in the relationship between peer relationships and delinquency in adolescence]. J Fam Better Life 2023;41:1-13. Korean.

10. Heo E, Nam J, Ko B, Kim J, Lee C, Choi K. [Mental health intervention for adolescents: a school-based program to address social anxiety]. Anxiety Mood 2018;14:88-98. Korean.

11. Kotyśko M, Frankowiak J. Psychometric characteristics of the center for epidemiologic studies depression scale-revised in a population of Polish university students. PLoS One 2023;18:e0290536

12. Ahn D. [Mental disorders in adolescents]. J Korean Med Assoc 2009;52:745-757. Korean.

13. Radloff LS. A self-report depression scale for research in the general population. Appl Psychol Meas 1977;1:385-401.

14. Cho MJ, Kim KH. Use of the center for epidemiologic studies depression (CES-D) scale in Korea. J Nerv Ment Dis 1998;186:304-310.

15. Eaton WW, Muntaner C, Smith C, Tien A, Ybarra M. Center for epidemiologic studies depression scale: review and revision (CESD and CESD-R). (3rd ed). In: Maruish ME, editor. The use of psychological testing for treatment planning and outcomes assessment. Mahwah: Lawrence Erlbaum Associates, 2004, p. 363-377.

16. Lee S, Oh S, Ryu SY, Jun JY, Lee K, Lee E, et al. [Validation of the Korean version of center for epidemiologic studies depression scale-revised(KCESD-R)]. Korean J Psychosom Med 2016;24:83-93. Korean.

17. Tsuboi H, Takakura Y, Tsujiguchi H, Miyagi S, Suzuki K, Nguyen TTT, et al. Validation of the Japanese version of the center for epidemiologic studies depression scale-revised: a preliminary analysis. Behav Sci (Basel) 2021;11:107

18. Cho MJ, Kim KH. [Diagnostic validity of the CES-D(Korean version) in the assessment of DSM-III-R major depression]. J Korean Neuropsychiatr Assoc 1993;32:381-399. Korean.

19. Heo EH, Choi KS, Yu JC, Nam JA. Validation of the center for epidemiological studies depression scale among Korean adolescents. Psychiatry Investig 2018;15:124-132.

20. Moon H, Oh K. [A validation study of the Korean social anxiety scale for children and adolescents]. Kor J Clin Psychol 2002;21:429-443. Korean.

21. Beidel DC, Turner SM, Morris TL. Social phobia and anxiety inventory for children: SPAI-C. North Tonawanda: Multi-Health Systems; 1998.

22. La Greca AM, Stone WL. Social anxiety scale for children-revised: factor structure and concurrent validity. J Clin Child Psychol 1993;22:17-27.

23. Yang J, Yang YR, Oh KJ. [Psychometric properties of the social anxiety scale for adolescents (K-SAS-A) in Korea]. Kor J Clin Psychol 2008;27:861-878. Korean.

24. Spielberger CD, Gorsuch RL, Lushene RE. Manual for the state-trait anxiety inventory (self-evaluation questionnaire). Palo Alto: Consulting Psychologists Press; 1970.

25. Cho SC, Choi JS. [Development of the Korean form of the state-trait anxiety inventory for children]. Seoul J Psychiatry 1989;14:150-157. Korean.

26. Skapinakis P. Spielberger state-trait anxiety inventory. In: Michalos AC, editor. Encyclopedia of quality of life and well-being research. Dordrecht: Springer, 2014, p. 6261-6264.

27. Reynolds WM. Suicidal ideation questionnaire. Odessa: Psychological Assessment Resources; 1987.

28. Orri M, Scardera S, Perret LC, Bolanis D, Temcheff C, Séguin JR, et al. Mental health problems and risk of suicidal ideation and attempts in adolescents. Pediatrics 2020;146:e20193823

29. Shin MS, Park KB, Oh KJ, Kim ZS, et al. [A study of suicidal ideation among high school students : the structural relation among depression, hopelessness, and suicidal ideation]. Kor J Clin Psychol 1990;9:1-19. Korean.

30. Kim H, Cho S, Kim J, Kang J, Shin M, Kim H, et al. [The effectiveness of mental health problems screening and treatment linkage in children & adolescents: community based study focused on ADHD and depression]. J Korean Acad Child Adolesc Psychiatry 2009;20:129-139. Korean.

31. Jung JW, Lee SH, Chun JY, Kim JE, Baek JW, Seo DW, et al. [A study on establishing a suicide prevention system through healthcare providers]. Sejong: Ministry of Health and Welfare, Korea Institute for Health and Social Affairs; 2010,20: p. 129-139. Korean.

32. Greenwald R. Child report of posttraumatic symptoms. Towson: Sidran Press; 1999.

33. Lee K, Jeong SH, Lee WK, Chung U. [Reliability and validity of the Korean version of the child report of post-traumatic symptoms (CROPS) and the parent report of post-traumatic symptoms (PROPS)]. J Korean Acad Child Adolesc Psychiatry 2011;22:169-181. Korean.

34. Rosenberg M. Rosenberg self-esteem scale (RSE) [Internet] Available at: https://integrativehealthpartners.org/downloads/ACTmeasures.pdf. Accessed July 5, 2024.

35. Bae H, Choi S, Yu J, Lee J, Choi K. [Reliability and validity of the Korean version of the Rosenberg self-esteem scale(K-RSES) in adult]. Mood Emot 2014;12:43-49. Korean.

36. Faul F, Erdfelder E, Buchner A, Lang AG. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods 2009;41:1149-1160.

37. Faul F, Erdfelder E, Lang AG, Buchner A. G^*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007;39:175-191.

38. Lee Y, Lee S, Kim C. [Korean adolescents’ weight control behaviors by BMI(body mass index) and body shape perception-Korea youth risk behavior web-based survey from 2010]. J Digit Converg 2012;10:227-232. Korean.

39. Angelakis I, Gooding P. Associations of anxiety and depression with suicide experiences in individuals with and without childhood trauma: the role of social support. Psychiatry Res 2022;309:114424

40. Anderson ER, Hope DA. A review of the tripartite model for understanding the link between anxiety and depression in youth. Clin Psychol Rev 2008;28:275-287.

41. Brodbeck J, Abbott RA, Goodyer IM, Croudace TJ. General and specific components of depression and anxiety in an adolescent population. BMC Psychiatry 2011;11:191

42. Van Dam NT, Earleywine M. Validation of the center for epidemiologic studies depression scale—revised (CESD-R): pragmatic depression assessment in the general population. Psychiatry Res 2011;186:128-132.

43. Kumar M, Jha AK. Adolescent brain development: limbic system and emotions. In: Shackelford T, editor. Encyclopedia of religious psychology and behavior. Cham: Springer, 2024, p. 1-10.

44. Redlich R, Opel N, Bürger C, Dohm K, Grotegerd D, Förster K, et al. The limbic system in youth depression: brain structural and functional alterations in adolescent in-patients with severe depression. Neuropsychopharmacology 2018;43:546-554.

45. Blakemore SJ, Choudhury S. Development of the adolescent brain: implications for executive function and social cognition. J Child Psychol Psychiatry 2006;47:296-312.

46. Park S. [A longitudinal study about relationship between somatic symptom and depression of adolescents]. J Digit Converg 2021;19:409-417. Korean.

47. Yu S. [A study on the structural relationship between perceived school climate, academic stress, depression/anxiety, and internet dependence in adolescents]. Korean Journal of Convergence Science 2024;13:157-176.

48. Kim BN, Park JH. [The effects of academic stress on depression in adolescents: the moderating role of response styles]. Korean J Child Stud 2013;34:113-128. Korean.

49. Mitchell RL, Phillips LH. The psychological, neurochemical and functional neuroanatomical mediators of the effects of positive and negative mood on executive functions. Neuropsychologia 2007;45:617-629.

50. Liew TW, Tan SM. The effects of positive and negative mood on cognition and motivation in multimedia learning environment. Educ Technol Soc 2016;19:104-115.

51. Roberts MB. Inventory of cognitive distortions: validation of a measure of cognitive distortions using a community sample (paper 325). Philadelphia: PCOM Psychology Dissertations; 2015.

52. Caballero A, Granberg R, Tseng KY. Mechanisms contributing to prefrontal cortex maturation during adolescence. Neurosci Biobehav Rev 2016;70:4-12.