Participants were recruited from four secondary schools in the Greater Bay Area of South China through convenience sampling between April and June 2023. All students studying in Secondary 1 (Grade 7) at the four schools were invited to participate in the study. They were informed that their participation was voluntary and confidential. Written consent forms were obtained before they began the questionnaire. Completing the questionnaire typically took 10–15 min. A total of 1,951 students participated in the study. After excluding responses that failed an attention check item and those with missing values, 1,196 (mean age = 13.07 ± 0.58) students were included in the data analyses (Table 1).

The present study was approved by the institutional review board of the authors’ university (HSEARS20230327004).

Data cleaning was conducted before the data analyses. To assess the students’ attention while completing the questionnaire, one attention item was included in the questionnaire. After excluding careless responses from the participants who did not answer the attention check correctly (N = 613) and those with missing values on the MLQ (N = 142), a total of 1,196 responses were retained. First, a three-step latent profile analysis was conducted to classify the participants’ MIL and to examine the associations between the different classes of MIL and sociodemographic factors. During the latent profile analysis, the following model fit indices were used: the Akaike information criterion (AIC), Bayesian information criterion (BIC), adjusted BIC (aBIC), entropy, and p-values for the Lo–Mendell–Rubin (LMR) test and bootstrapped likelihood ratio test (BLRT). Smaller values of the ACI, BIC, and aBIC indicate a better model fit. An entropy value closer to 1 indicates higher classification accuracy. p-values of less than 0.05 for the LMR test and BLRT indicate a satisfactory model fit (Asparouhov and Muthén, 2014). Furthermore, multivariate analysis of variance (MANOVA) was conducted to compare mental health (i.e., depression, anxiety, and stress) across the different MIL profiles. All data were analyzed using Mplus Version 8 and SPSS Version 23.0.

As self-report measures may produce common method bias (Zhou and Long, 2004), we first used Harman’s one-factor test to assess for common method bias. For the MILQ and DASS-Y assessments, a single factor accounted for 37.72% and 48.55%, respectively, which is considered acceptable for further analysis. The participants scored higher on the PMIL than the SMIL, and their stress levels were higher than anxiety and depression (Table 2).

For the latent profile analysis of MIL, the present study initially started with two classes and then increased the number of classes by one each time. The results showed that the AIC, BIC, and aBIC were decreasing. The p-value for the LMRT was greater than 0.05 when there were four classes. The percentage of class 1 was less than 5% when there were five classes. Therefore, overall, a three-class solution was deemed the best, despite having a lower than the four-class and five-class solutions (Table 3).

Class 1 represented a medium level of PMIL and SMIL and was therefore labeled “medium MIL.” Approximately half of the participants (n = 600, 50.04%) were searching for their MIL at a medium level and found a medium level of the presence of MIL. Class 2 represented a lower level of PMIL and SMIL compared to the other two classes, with the smallest percentage (11.59%, n = 139). The PMIL was lower than the SMIL, leading to its label as “low and searching MIL.” Within this class, the participants scored higher on the subscale of the SMIL and lower on the subscale of the PMIL. The remaining participants (n = 460, 38.37%) were categorized into Class 3 and scored higher on both PMIL and SMIL. Thus, Class 3 was labeled “high MIL.” Taking the three classes together, the differences (Table 4; Figure 1) in the PMIL were larger than those in the SMIL. This suggested that the participants were searching for MIL at a similar level; however, some of them found high MIL (Class 3), half of them enjoyed medium MIL (Class 1), while a small percentage of them had low MIL (Class 2).

Regarding the sociodemographic factors associated with the MIL profiles, the results revealed that the participants whose mothers had higher education levels and those from intact families were 1.30 times and 3.21 times more likely, respectively, to be classified into Class 3 (high MIL) compared to Class 2 (low and searching MIL; Table 5).

Based on the class membership that resulted from the latent profile analysis, this study examined whether the participants in the different classes experienced different mental health statuses, including depression, anxiety, and stress. The results revealed that the participants in the three classes reported significantly different levels of depression, anxiety, and stress. The participants in Class 3 (high MIL) reported the lowest levels of depression, anxiety, and stress, while those in Class 1 (medium MIL) reported moderate levels and those in Class 2 (low and searching MIL) reported the highest levels. The most significant difference was in depression (χ² (2) = 69.75, p < 0.001; Table 6).

In the present study, the participants generally had a good sense of meaning in life. Three latent classes were identified, namely “high MIL,” “medium MIL,” and “low and searching MIL.” This finding is largely consistent with Ma et al.’s (2023) study, in which high school students (mean age = 16.33 ± 1.01) were classified into four classes: “high MIL,” “medium MIL,” “searching MIL,” and “low MIL.” Both Ma et al.’s (2023) study and the present study classified Chinese adolescents based on their levels of MIL. Moreover, adolescents with medium and high levels of MIL accounted for a large percentage in both studies, indicating that the majority of adolescents in China reported medium and high levels of MIL.

Both studies exhibited a pattern of “high SMIL and low PMIL” within a single class: the “searching MIL” class in Ma et al.’s (2023) study and the “low and searching MIL” class in the present study, with the gap between high SMIL and low PMIL being smaller. The two studies revealed a similar phenomenon: Chinese adolescents were searching for MIL, but not everyone had found it. Unlike their peers in the “low MIL” class in Ma et al.’s (2023) study, who scored the lowest in both PMIL and SMIL groups, these adolescents were actively seeking something that would make their lives meaningful and significant, as well as trying to figure out their life purposes; however, they failed to find them. They may need external support in the process of searching for their MIL. For example, secondary schools could provide a life education curriculum to help these adolescents understand themselves and their life events, as well as explore their purpose and the meaning of their lives.

However, the pattern “high SMIL and low PMIL” was not reported in He et al.’s (2024) study conducted on college medical students (mean age = 21.73 ± 1.13) in China, which identified three classes through an LPA, and the PMIL was not lower than the SMIL. This may suggest that the phenomenon of adolescents searching for but failing to acquire MIL disappears when they enter early adulthood. Another possible reason is that He et al. (2024) conducted their study among medical students who might have encountered more opportunities to reflect on MIL during their studies (e.g., interacting with patients). This indicates that adolescents who are still in their searching phase may need more opportunities to be inspired by MIL.

The present study revealed that adolescents from intact families were more likely to be classified in the high MIL class compared to being classified in the low and searching MIL classes. This finding is consistent with the results of Shek et al.’s 2021 study, which reported that family intactness positively predicted MIL among adolescents. One possible explanation is that adolescents from intact families may have more resources and fewer challenges in their search for and acquisition of MIL. The attention and resources from both biological parents can help adolescents gain more life experiences, which is essential for searching for and acquiring MIL. This process requires life experience and the ability to interpret and make sense of what has happened in our lives (Steger, 2009). Parental support is related to MIL even among young adults through optimism and identity commitment (Kealy et al., 2022). On the contrary, adolescents from single-parent families and stepfamilies reported more suicidal thoughts (Garnefski and Diekstra, 1997), which may indicate a failure in searching for and acquiring their MIL.

The present study also found that the adolescents whose mothers had higher education levels were more likely to be classified in the high MIL class compared to being classified in the low and searching MIL class. One possible reason is that mothers with higher education levels may offer more opportunities for their children to explore the world and themselves, which are important for shaping one’s MIL. As for why mothers’ education levels, rather than fathers’, are related to classification in the high versus low and searching MIL classes, it may be because mothers play a larger role in childcare and parenthood (Shek et al., 2019; Zou et al., 2019). Therefore, mothers’ education levels play a more significant role than those of fathers.

The findings regarding the importance of both family intactness and mothers’ education levels are consistent with the literature review, which identifies the family as the most important source of meaning in people’s lives (Glaw et al., 2017).

The present study revealed that MIL was generally related to adolescents’ depression, anxiety, and stress. The adolescents with high MIL reported significantly less depression, anxiety, and stress. This finding is consistent with previous studies among adolescents in the US (Dulaney et al., 2018), Spain, and Russia (Shub et al., 2024), highlighting the cross-cultural importance of MIL as a protective factor for adolescents’ mental health. Moreover, this finding builds on Krok (2018), which demonstrated that MIL is positively correlated with subjective wellbeing and psychological wellbeing, including factors such as autonomy, environmental mastery, personal growth, positive relations with others, purpose in life, and self-acceptance. While Krok (2018) revealed a positive relation between MIL and the positive indicators of wellbeing, the present study revealed a negative relation between MIL and negative mental health issues, such as depression, anxiety, and stress. Krok (2018) also suggested that the purpose embedded in the concept of MIL appears central to the formation of wellbeing in late adolescents, whose mean age is approximately 18 years in Poland. This study also found that the central role of MIL in influencing mental health is true for early adolescents (age = 13.07 ± 0.58) in China. Overall, these findings indicate the importance of leading a meaningful and purposeful life in fostering good mental health and wellbeing holds true across different cultures and throughout both early and late adolescent years.

Another noteworthy observation is that the largest difference among the three categories of MIL was observed in depression rather than in stress or anxiety. This highlights the crucial role of MIL in addressing depression among adolescents. Finding out the purpose and meaning of their lives could protect adolescents from depression, a major health concern for them (Daly, 2022). Therefore, educators and social workers are encouraged to provide education, interventions, and support in the search for MIL to prevent depression among adolescents.

Although the relationship between MIL and mental health has been examined before (for a literature review, see Glaw et al., 2017; Li et al., 2021), this study makes three unique contributions:

First, it employed a person-centered method instead of a variable-centered method. As mentioned earlier, searching for and acquiring MIL are individual experiences and endeavors. Therefore, a person-centered approach may be more appropriate for capturing the differences among individual adolescents. For instance, the latent profile analysis revealed a combination of relatively high SMIL and low PMIL in Class 2, which may not be easily identified using a variable-centered approach.

Second, the results revealed that the differences in mental health, such as depression, anxiety, and stress, were mainly caused by the PMIL rather than the SMIL among adolescents. This highlights that high SMIL does not necessarily guarantee high PMIL. Therefore, the intervention program should focus on helping adolescents find their MIL and identify a life purpose. For example, secondary schools can provide life education courses to enhance students’ understanding of life events, help them understand themselves, and encourage them to explore the spirituality of life.

Finally, this study was conducted in the aftermath of the COVID-19 pandemic, and the results indicated that even after a 3-year-long global health crisis, the majority of Chinese adolescents possess a relatively medium to high MIL.

The study has several limitations that should be recognized. First, as it was a cross-sectional survey, the adolescents with better mental health might have had more mental resources to focus on deep questions regarding the meaning and purposes of their lives, making them more likely to find their meaning in life. Second, the convenience sampling might have limited the generalizability of the findings of this study. In the future, including adolescents from different regions in China in the survey may enhance its representativeness. More importantly, while we gathered snapshot data on the SMIL and PMIL, searching for and acquiring MIL are not endpoint statuses. Understanding the processes of SMIL and changing PMIL among adolescents may yield more interesting findings.