Post-earthquake resilience across generations: a CD-RISC-based empirical study

Indonesia is among the nation’s most vulnerable to disasters due to its geographic location, socioeconomic vulnerabilities, and other cultural and political considerations (Soetanto et al., 2020). Natural disasters are closely related to poverty, creating a cyclical relationship where poverty increases vulnerability, and vulnerability exacerbates poverty (Ellis, 2006). Disasters exacerbate poverty and social ills (Martins et al., 2020), as individuals with limited resources struggle to recover, affecting their overall resilience (Briguglio et al., 2009). Disasters also disrupt economic growth and hinder efforts to escape poverty (Amaratunga et al., 2018), with poor families particularly vulnerable (Bui et al., 2014; Boubacar et al., 2017). A key consequence of post-disaster events is the psychological and behavioral impact on survivors, with younger generations in particular often experiencing trauma and stress that undermine their personal resilience (Mabruri, 2009).

Bantul Regency was heavily impacted by past earthquakes, causing long-term socio-economic disruptions across demographics, economy, health, education, and hazard-prone areas (Raharjo et al., 2006). For generations Y and Z, the disaster brought psychological trauma, anxiety, and weakened resilience, worsened by slow recovery and infrastructure damage (Masykur, 2006; Mabruri, 2009). These conditions led to significant youth outmigration—22% of young people in Jetis District now work outside the area (Hizbaron et al., 2012; Bantul, 2024)—and contributed to poverty due to low human capital and limited local engagement (Feriyanto, 2015). While earlier studies examined survivor behavior (Nindyati, 2012) and cultural adaptation (Christiani and Ikasari, 2020), none have focused on how Y and Z generations use personal resilience to support family livelihoods. This study fills that gap by investigating four villages in Jetis District—Trimulyo, Sumber Agung, Patalan, and Canden—as a case study of youth resilience in post-disaster rural Indonesia.

In this context, Generation Y (born 1981–1997) and Generation Z (born 1997–2012) make up the majority of Indonesia’s productive-age population and will play a critical role in the nation’s development in the digital era (Ekananda and Rachma Marcillia, 2019; Central Bureau of Statistics (BPS), 2020). However, in vulnerable areas, human capital quality remains low, and the resilience of these generations is challenged by issues such as low capability, misinformation, value erosion, poor emotional control, consumptive tendencies, weak ethics and skills, and unproductivity (Prabowo and Putranta, 2016; Dwidienawati and Gandasari, 2018; Nurhajati et al., 2018; Rahayu and Nurfauziah, 2020). On the other hand, personal resilience plays a vital role in strengthening social, human, and economic capital, as well as community resilience overall (Bastaminia et al., 2017).

This study aims to examine the personal resilience of generations Y and Z who represent 70% of the productive-age population in Jetis District, Bantul Regency, Indonesia—a district that remains one of the poorest in the region. By exploring how these generations rebuild their economic independence and contribute to family resilience after the disaster, this research provides new insights into the post-disaster recovery process and human capital development.

Personal resilience refers to an individual’s ability to recover from adversity, restore well-being, and maintain physical, mental, and emotional health (Smith et al., 2013; Hung and Appleton, 2016; Satici et al., 2021). Post-earthquake conditions require knowledge of generation characteristics in terms of the ability to survive, mitigate, and recover (Diržytė et al., 2017). Richardson’s (2002) three-wave theory outlines resilience as: (1) the ability to bounce back from adversity; (2) using challenges as opportunities for growth; and (3) reintegrating life disruptions with renewed strength and wisdom.

Several validated instruments exist for measuring personal resilience, each with distinct theoretical foundations and applications, as shown in Table 1. The Brief Resilience Scale (BRS) focuses primarily on the speed of recovery from stress through a concise six-item assessment (Smith et al., 2008). While useful for measuring immediate stress responses, its scope may be too narrow for capturing the complex, long-term adaptation required in post-disaster recovery (Ahern et al., 2006).

On the other hand, the Resilience Scale for Adults (RSA) takes a broader approach by assessing both interpersonal and intrapersonal resources, including self-perception and social support networks (Friborg et al., 2003; Rossi et al., 2021). Although comprehensive, this scale was primarily developed for general populations and high-risk groups, potentially lacking specificity for trauma-related resilience assessment.

Another approach is represented by the Ego-Resiliency Scale (ER89), which originates from personality psychology (Block and Kremen, 1996). This instrument emphasizes traits such as ego control, intellectual functioning, and openness to experience. While valuable for understanding personality-based adaptation, it may not fully capture the specific challenges of post-trauma recovery (Chen et al., 2023).

In contrast to these more narrowly focused instruments, the Connor-Davidson Resilience Scale (CD-RISC) offers a comprehensive framework for assessing resilience, capturing both adversity resistance and the positive personality traits linked to resilience (Connor and Davidson, 2003). High CD-RISC scores reflect strong coping skills and emotional regulation, fostering constructive behaviors in the face of challenges (Werner and Smith, 1993). Aligned with contemporary resilience theories (Richardson, 2002) and proven reliable across demographics, the CD-RISC is especially suited for examining generational differences between Generation Y (1981–1996) and Generation Z (1997–2012), who may have experienced similar disasters at different developmental stages (Diržytė et al., 2017). By integrating psychological, social, and spiritual dimensions (Henfrey, 2018), the CD-RISC enables a multidimensional understanding of resilience’s role in post-disaster recovery and economic self-sufficiency. Accordingly, this study employs the CD-RISC for its ability to measure adaptability, stress tolerance, and acceptance of change—key to analyzing resilience across generations (Chen et al., 2023).

In terms of recognizing Generation Y and Generation Z based on Generation Theory through grouping generations based on the year of birth, the grouping is as follows: those born before 1946 are called the traditional generation, baby boomers were born between 1946–1963, and generation X was born between 1964–1979 (Berenda and Mannheim, 1953). The generation grouping was further developed by Spiro (2006) and Schroth (2019), who added Generation Y (born up to 2000) and Generation Z (born up to 2013).

Generation Y, known as the millennial generation, is characterized by confidence and an emphasis on values in work motivation (Ryan and Deci, 2000; Srivastava, 2016). This generation is reinforced by Cennamo and Gardner (2011) as having an internet technology-based lifestyle, appreciating changing values, being adaptable and tolerant, flexible, able to work in teams (Dwyer, 2009), and proficient at multitasking (Helyer and Lee, 2012).

Generation Z is generally an educated generation that uses technology to solve problems (The Adecco Group, 2020). They believe that work influences social status and they always receive support and protection from their parents (Philip and Garcia, 2013; Gomez, 2022). According to Dolot (2018), Generation Z hopes for fast feedback, instant careers, and multitasking, and organizes all activities using calendars and applications.

In terms of resilience, several studies have raised concerns. During the COVID-19 pandemic, a study revealed Generation Z exhibited lower resilience than Generation X despite higher openness to change and self-enhancement, along with more positive attitudes toward flexible learning, while showing no intergenerational differences in flexible work or online consumption attitudes (Harari et al., 2023). These findings align with CD-RISC-25 assessments showing Generation Z’s particular struggles with adversity adaptation, where pandemic-induced social isolation exacerbated loneliness and eroded self-efficacy (Antapurkar and Choudhari, 2024). These generational differences underscore the need to examine the unique socio-cultural and psychological factors that influence resilience across cohorts. Understanding these patterns is essential for contextualizing the resilience of young adults—particularly in vulnerable or disaster-affected settings like rural Indonesia—as explored in this study.

On May 27, 2006, a 6.3-magnitude earthquake struck the south-central coast of Java, Indonesia, causing severe destruction in the Special Region of Yogyakarta (DIY). The disaster killed 5,744 people and damaged around 628,000 homes (World Bank, 2012), including 18,119 small businesses in Bantul Regency—56% of which were owned by women—exacerbating poverty and vulnerability (World Bank, 2005).

Within Bantul Regency, the Jetis Sub-district was identified as one of the hardest-hit areas during the earthquake (Raharjo et al., 2006), comprising four villages—Trimulyo, Sumber Agung, Patalan, and Canden—with about 58,550 residents. Its high seismic risk and low-income, low-education communities (Suryanto and Kuncoro, 2012) have hindered infrastructure development, driving many to seek jobs elsewhere.

Nearly 49% of Jetis’s population belongs to Generations Y and Z (BPS, 2020). Despite being in their most productive years, many still struggle with economic instability. Jetis remains one of the poorest sub-districts in Bantul, reflecting unequal recovery outcomes. While Bantul’s overall unemployment dropped from 9% to 5% and poverty from 30.1% to 14.2% in the decade after the quake (BPS, 2020), small and medium-sized enterprises (SMEs) in Jetis faced ongoing challenges, with many failing to recover. Although the 2006 Yogyakarta earthquake occurred nearly 2 decades ago, its long-term effects remain significant, particularly in the context of human development and economic recovery. This indicates that despite broader economic gains, localized vulnerabilities persist and have deepened over time.

Furthermore, Hoddinott et al. (2013) found that major disasters like earthquakes can have lasting effects on childhood development and education—key elements of human capital. In Jetis Sub-district, one of Bantul’s poorest areas (BPS, 2020), Generations Y and Z were in early childhood during the 2006 earthquake, a critical stage for development. Today, they face persistent issues such as low education, high youth unemployment—especially among women—and limited income opportunities. This study suggests that the earthquake may have contributed to these outcomes, highlighting the need to assess personal resilience to guide future development efforts.

Understanding Jetis’s demographic and socioeconomic conditions is crucial for framing this study. The large proportion of youth—especially from Generations Y and Z—represents untapped potential for community development. High personal resilience is linked to improved behavior, ethics, and problem-solving under adversity. Many resilient individuals exhibit perseverance, motivation, and adaptability through digitalization and multitasking, making them key drivers in post-disaster recovery and future growth.

This study employed a quantitative approach to investigate the personal resilience of Generations Y and Z in a post-earthquake context. The primary instrument used for measuring resilience was the Connor-Davidson Resilience Scale (CD-RISC), which comprises 25 items designed to capture various indicators of personal resilience. The instrument employed a 5-point Likert scale ranging from 0 to 4 (Likert, 1932). Based on the scoring criteria adapted from Connor and Davidson (2003), resilience levels were categorized as follows: very bad (<30), average (30–70), good (70–90), and excellent (90–100).

To ensure the instrument’s suitability for the target population, a rigorous multi-step translation and adaptation process was followed, drawing from Beaton and Guillemin (2000) and Cruchinho et al. (2024) (see Figure 1). Initially, the original English version of the 25-item CD-RISC was translated into Bahasa Indonesia by bilingual experts, with their versions subsequently synthesized through consensus discussions. A separate team of blinded bilingual translators then performed back-translation to verify conceptual equivalence. All versions were reviewed in a harmonization step to resolve discrepancies and ensure both semantic and conceptual equivalence.

Beyond linguistic translation, the 25 original items were systematically mapped into 7 locally relevant resilience indicators that reflect both the CD-RISC’s core constructs and the lived experiences of disaster-affected communities in Yogyakarta. As shown in Table 2, this adaptation process created meaningful clusters of items with shared local significance - for instance, original items like “Tend to bounce back after illness or hardship” (#8) and “Things happen for a reason” (#9) were grouped under the indicator “Possesses the strength and experience to overcome difficulties, failures, and heartbreaking losses” (X.5.1.2). This culturally grounded mapping ensured the instrument’s face validity while maintaining the theoretical foundations of resilience as conceptualized in the original CD-RISC.

A pilot test was conducted with 20 participants demographically similar to the study sample to evaluate clarity and contextual fit. Based on this feedback, minor revisions were incorporated. The refined instrument was then field-tested with 591 participants, representing a subset of 1,182 valid responses collected across four villages in Jetis District, Bantul Regency: Trimulyo, Sumber Agung, Patalan, and Canden. A cluster sampling strategy was employed to ensure proportional representation from each village, as shown in Table 3.

The psychometric validation phase included assessments of reliability and validity using SmartPLS, involving key indicators such as Cronbach’s Alpha, Average Variance Extracted (AVE), discriminant validity, and Confirmatory Factor Analysis (CFA). As the study focused solely on a Javanese cultural context without cross-cultural comparison, formal cross-cultural validation was not required. Instead, the analysis aimed to interpret resilience as defined by CD-RISC within this specific socio-cultural setting.

Quantitative analysis with SPSS included descriptive statistics, correlation, and reliability tests, while SmartPLS supported advanced model validation. Resilience scores were compared by generation, gender, and village. Qualitative data—gathered through field observations, interviews with village leaders, and analysis of development documents—added contextual depth. This mixed-methods approach, supported by culturally grounded instruments and triangulation, offered a comprehensive view of generational resilience post-disaster.

The following findings are derived from qualitative data collected through in-depth interviews and field observations conducted across four villages. This section presents participants’ perspectives and lived experiences regarding demographic characteristics, socioeconomic conditions, and post-earthquake resilience.

In Trimulyo, Generation Y comprises 68.7% of the population, with male dominance nearly double that of females. Among them, 80% are married, 38% of women do not work, and 14% of Generation Y is unemployed. Education levels are low due to financial constraints. In Generation Z, 52% are still in school and unmarried; 41% of males are already married and working. While nearly all use digital devices, most work in informal sectors with incomes between IDR 1.2–5 million. Government aid supports 16.5% of residents. Most respondents experienced the 2006 quake, with some families affected by injury or death. Community rebuilding efforts have since improved resilience.

Similarly, in Sumber Agung Village, Generation Y makes up a larger portion of the population at 55.6%, compared to Generation Z at 44.4%, with Generation Z predominantly female and twice as many as males. Around 85% of Generation Z and 43% of both generations are unemployed, mainly women. Workers earn unstable incomes, often below minimum wage. While tech use is high for communication, few use it for income—except 41.7% of Generation Y, who use social media for business. Education levels are low, especially among Generation Z. During the earthquake, Generation Y generally stayed calm and supported others.

In Patalan Village, the population composition is almost three times more Generation Y than Generation Z, potentially boosting development during this decade. Within Generation Y in Patalan Village, 69.5% are women, whereas Generation Z has a more balanced gender distribution with 45% men and 55% women. Despite the high number of Generation Y residents, the unemployment rate is quite high at 34%, as many women in the village are housewives. This presents a significant challenge for improving welfare, as many productive-age individuals are engaged in household chores, educating children, and managing the home, without generating income, thus remaining dependent on other productive age members.

Generation Y in Canden Village consists of 49.3% women, and Generation Z is composed of 50% women. This results in a more balanced gender distribution in Canden Village compared to other areas. Generation Y’s unemployment rate is low (14.5%), mostly among housewives—70% of whom still earn income. Both generations actively adopt technology for small businesses, particularly in trade and services. Around 23% receive social support. Many use digital platforms for online stores, culinary sales, and services. Innovations in crafts and processed foods are prominent. Among Generation Z, 65% are in school and 35% already working, indicating more favorable socioeconomic dynamics than other villages.

This section presents an integrated analysis of the measurement reliability and validity of personal resilience across the four studied villages—Trimulyo, Sumber Agung, Patalan, and Canden—alongside the correlation between personal resilience (PR) and generation (GEN). Using SmartPLS, key psychometric metrics such as AVE score, discriminant validity, Pearson validity, and Cronbach’s Alpha were analyzed to ensure the robustness of the resilience construct across diverse village settings (Table 4). High AVE scores in all villages—led by Patalan (0.982) and Trimulyo (0.874)—indicate that traits like adaptability, self-confidence, and problem-solving accurately represent personal resilience.

However, discriminant validity scores reveal a notable overlap between resilience and broader life factors, such as education, financial security, and community support. For instance, Patalan’s low discriminant value (0.166) suggests that resilient individuals also tend to benefit from strong social and economic foundations. This overlap underscores the complexity of resilience as a construct, which, while rooted in personal traits, is often reinforced by external conditions. The Pearson validity and Cronbach’s Alpha scores in all four villages further confirm that the survey instruments were well understood and produced consistent, reliable responses.

In simple terms, the statistical findings indicate that the measurement of personal resilience works well across the four villages. The high scores suggest that respondents understood the concept clearly and answered reliably. Among the four villages, Patalan shows the most consistent and accurate measurement of resilience, possibly reflecting a strong community identity or shared values that reinforce resilient behavior.

The study also investigates the correlation between personal resilience and generational status (Generation Y and Z) across the four locations are shown in Figure 2. While the overall correlation is weakly positive, village-specific insights reveal critical differences. Trimulyo, with the highest correlation (0.234), shows that in more economically stable settings, resilience is more likely to translate into generational development. Here, resilience is characterized by practical adaptability and strong community networks, supported by higher incomes and low dependence on government aid. This suggests that when support systems are stable and education levels moderate to high, resilience becomes a tool for generational progress.

In contrast, Sumber Agung—despite high employment among Generation Y—demonstrates minimal influence from challenge-driven resilience traits due to limited upward mobility. Here, the role of education and employment is constrained by systemic barriers, such as stagnant wages and lack of capital access, which blunt the transformative power of personal resilience.

Patalan presents a unique case, where spiritual values—such as trust in God, self-control, and moral responsibility—serve as the foundation for resilience. These values are deeply embedded in the local culture and provide a powerful internal support system. Meanwhile, Canden reflects a resilience profile focused on persistence and self-belief, shaped by recovery from past adversities like the 2006 earthquake. While external aid in Canden provides short-term relief, it also risks reducing the drive for long-term self-reliance.

Taken together, the findings illustrate that while personal resilience is consistently measurable and reliable across villages, its practical impact is heavily mediated by education, employment opportunities, and community-level support systems. Education enhances cognitive and emotional adaptability; employment offers purpose and stability; and support systems—both formal (government aid) and informal (family, faith, community)—amplify the ability to recover and thrive. Therefore, fostering resilience in younger generations requires not just cultivating internal strengths, but also building external environments that enable those strengths to flourish.

The Table 5 presents the personal resilience scores of Generation Y and Generation Z, segmented by gender, across four villages: Trimulyo, Sumber Agung, Patalan, and Canden. The findings highlight both quantitative disparities and qualitative patterns, underscoring the influence of structural and contextual variables in shaping resilience development.

The data indicates that Canden consistently exhibits the highest personal resilience scores across all categories, with values reaching 86.5 (Male Y), 84.9 (Female Y), and 87.0 (both Male Z and Female Z). This suggests that individuals in Canden, regardless of generation or gender, demonstrate stronger resilience compared to other villages. Conversely, Patalan consistently shows the lowest resilience scores, particularly among Generation Z, where Male Z scores 60.3 and Female Z scores 61.0. This marked disparity suggests that environmental, social, or economic conditions in Patalan may play a role in diminishing resilience levels, particularly among younger individuals.

Comparing generations, Generation Y (both male and female) consistently reports higher resilience scores than Generation Z across all villages. The average scores further confirm this trend, with Generation Y males averaging 79.5 and females averaging 81.5, while Generation Z males and females score lower at 76.4 and 75.1, respectively. This suggests that older individuals may have had more opportunities to develop coping mechanisms and resilience, possibly due to greater life experience or stability.

Within each generation, gender differences are also evident. Female Y (average 81.5) scores slightly higher than Male Y (average 79.5), indicating a marginal advantage in resilience among women in Generation Y. However, in Generation Z, Male Z (average 76.4) scores slightly higher than Female Z (average 75.1), though the gap is less pronounced. These findings highlight potential gender-related variations in resilience development across generations.

Interestingly, while resilience scores for Generation Y are relatively stable across villages, Generation Z exhibits more drastic fluctuations. This variability suggests that younger individuals are more sensitive to external factors such as economic conditions, education, and community resilience programs. If such external factors are not well-managed, Generation Z may struggle to develop resilience at the same level as Generation Y.

The findings of this study offer significant insights into the development of personal resilience among Generations Y and Z in post-disaster contexts. Although the 2006 Bantul earthquake serves as a historical backdrop, resilience appears to be shaped by a broader constellation of factors—psychosocial dynamics, socio-economic structures, community environments, and life experiences—rather than by generational identity alone.

Nearly twenty years after the disaster, the earthquake’s long-term consequences continue to influence resilience development through restricted educational access, scarce employment opportunities, and cycles of inherited poverty. The adapted application of the CD-RISC in this Javanese rural setting demonstrates how fundamental resilience components—particularly adaptability and purpose—take distinctive forms within this context. Two dimensions stand out as particularly underdeveloped: emotional regulation appears shaped by local collective norms governing emotional expression, while future orientation remains constrained by structural barriers like economic instability and limited schooling rather than individual shortcomings alone. This contextual adaptation confirms generational labels are insufficient; resilience must be interpreted through intersecting factors like psychosocial dynamics, economic conditions, educational levels, employment status, and life experiences.

The results indicated that Generation Z consistently demonstrated lower resilience than Generation Y across all surveyed villages. This generational disparity aligns with findings from previous studies using CD-RISC, where Generation Z was found to have lower capacity for adapting to adversity, coupled with heightened vulnerability to stress and emotional instability—especially in the wake of the COVID-19 pandemic (Antapurkar and Choudhari, 2024). Additional research by Harari et al. (2023) also found that although Generation Z showed greater openness to change and more positive attitudes toward flexible learning, their resilience scores remained comparatively low. These findings suggest that while Generation Z may embrace change and innovation, their foundational coping mechanisms remain less developed, particularly in contexts of economic or emotional strain.

Psychosocially, Generation Y benefits from having more time to develop coping skills and stable identities, while Generation Z—still navigating identity formation amid economic instability—requires targeted support such as mentoring and peer groups to build emotional regulation, a key capacity for their important role in post-disaster recovery. Economically, better-resourced villages like Canden demonstrate the positive impact of educational infrastructure and youth engagement initiatives, whereas less-resilient areas like Patalan face limited employment opportunities, weak educational support, and digital divides that hinder economic adaptation—especially for young women and informal workers excluded from the formal sector. To address these disparities, resilience-building efforts must prioritize structural interventions, drawing from models like Canden with strong local networks and intergenerational solidarity, and supported by policies that integrate resilience education, expand digital literacy, promote gender-equitable employment, and provide accessible mental health services in under-resourced communities.

Future research should move beyond the Generations Y and Z framework. Comparative analyses incorporating older generational cohorts and cross-cultural samples could provide enhanced understanding of resilience dynamics, particularly considering the culturally contingent interpretations of CD-RISC dimensions. Longitudinal research tracking resilience evolution is also essential. Shifting focus from generational labels to the intersectional and structural determinants identified here lays the groundwork for adaptable resilience strategies in diverse disaster-affected communities.

This study examines the personal resilience of Generations Y and Z across four villages—Trimulyo, Sumber Agung, Patalan, and Canden—by analyzing their demographic characteristics, socioeconomic conditions, and experiences with past adversities, particularly the 2006 earthquake. This study highlights that personal resilience among Generations Y and Z in post-disaster contexts is not determined solely by generational identity, but is instead shaped by a complex interplay of psychosocial dynamics, economic conditions, educational levels, employment status, and life experiences.

Across all villages, Generation Y demonstrates a higher level of resilience, attributed to their greater life experience, problem-solving abilities, and responsibility in supporting their families. In contrast, while Generation Z benefits from better educational attainment and technological adaptation, they face challenges in economic independence and workforce participation. The study also highlights significant variations in resilience across villages, with Canden Village showing stronger economic adaptability through technology-based businesses, while other villages struggle with higher unemployment and dependency on social assistance.

The role of past experiences, particularly the 2006 earthquake, plays a crucial part in shaping resilience. However, it is the ongoing support systems, education, and employment opportunities that enable communities to translate adversity into growth. Family networks, cooperatives, and mentoring programs offer emotional and practical support, while vocational and digital training equip individuals with skills for income diversification and economic stability. For example, digital literacy workshops in Canden helped youth launch online businesses, directly boosting resilience. Locally relevant jobs, such as agricultural services in Patalan or eco-tourism in Trimulyo, further enhance problem-solving and financial security.

These findings have important implications for both theory and practice. Theoretically, this study validates the CD-RISC while enriching it with qualitative insights to better understand how resilience develops among Generations Y and Z in post-disaster rural contexts. Practically, it highlights the importance of skill-based education, digital programs, and locally adapted employment to build lasting resilience and economic independence. By identifying the distinct resilience patterns in each village, policymakers and community stakeholders can formulate strategies that foster sustainable development and economic independence for future generations.