Measures

ACEs were assessed using the Chinese version of the Adverse Childhood Experiences International Questionnaire (ACE-IQ) [51, 52], which has been established to show good reliability coupled with validity in China [53]. The 32-item measure explores exposure to four domains of ACEs: neglect, family dysfunction, abuse, and violence. Each domain includes several categories. We used 28 items from all categories, with the exception of exposure to collective violence/war (four items). Participants were asked to respond to the 28 questions on the basis of their encounters during their first 16 years of life. Each item was then constructed as a dummy variable with a score of “0” assigned if the participant had never experienced it and a code of “1” assigned if the participant had experienced it during the first 16 years of life. The overall number of ACE items to which the subject was “exposed” was summed to construct an ACE score ranging between 0 and 28. This study focused on the findings of subjects’ overall ACE exposure (ACE score). Overall, the Cronbach’s alpha coefficient of the ACE-IQ (ACE score) was measured as 0.84.

RNLEs were investigated by the Adolescent Self-Rating Life Events Checklist (ASLEC) [31, 54], which is utilized to explore the impact of negative life events on teenagers (including undergraduates) over the previous 12 months. The ASLEC includes 26 items linked to five dimensions: study stress, punishment, relationship pressure, loss, and healthy adaptation. The checklist uses a six-point scale (1 = no negative life events to 6 = negative life events with serious impact). The higher a score is, the more distressing the associated negative life event was. In this study, the Cronbach’s alpha coefficient of the scale was measured as 0.92.

Self-efficacy to avoid NSSI (NSSI-SE) was explored via an 11-item questionnaire adapted from the scale measuring NSSI-SE to avoid committing non-suicidal self-injury [43]. Participants reported on a 5-point scale ranging from 1 (very unfit) to 5 (very fit) whether they believed they could resist engaging in NSSI in certain situations based on statements such as “I feel like I can accept myself,” “I need to relieve inner pressure or need a release,” and “I have easy access to (self-harming) tools.” The scale included three subscales, for “confidence,” “emotion regulation temptations,” and “reinforcing temptations.” The greater the total score is, the stronger a student’s level of NSSI-SE to resist NSSI is. The Cronbach’s α of the current sample was measured as 0.89.

NSSI was tested using the 18-item Adolescents Self-Harm Scale (ASHS) [55]. This instrument explores the degree to which persons repeatedly or deliberately self-harm via diverse methods (e.g., hitting, skin-cutting, or stabbing) without harboring suicidal intentions. Two dimensions of each of the 18 items were evaluated. The first evaluated the number of times a participant deliberately adopted this method to self-harm at any time previously (frequency) and used a four-point scale (0 = has not happened, to 3 = has happened more than five times); the second item evaluated the extent to which a previous act had hurt the participant’s body at any time (degree) on a five-point scale (0 = no harm to the body, to 4 = extreme severe harm to the body). Each item’s score was equal to the frequency multiplied by the extent, and the overall score (summed score across all items) was employed to compute the overall extent of NSSI. The higher a participant’s total score was, the higher their degree of NSSI was. The Cronbach’s α for ASHS was measured as 0.92.

Finally, subjects were asked to report their age, gender, grade, and study major and their parents’ education level (6 levels from 1 = elementary school graduate or below to 6 = postgraduate degree graduate or above). Subjective socioeconomic status was explored via the MacArthur ladder scale of subjective social status (10 steps) [56].