Cumulative family risk
Theoretically, all ecological factors can be included in the measurement of cumulative risk. However, in terms of the necessity and feasibility of the study, only important risk factors that are closely related to development outcomes should be included [14]. In the research design stage, we systematically searched and read the literature on non-suicidal self-injury in families or parents and adolescents in the past five years, and followed the following principles to screen for family cumulative risk factors: (1) Systematic: Under the guidance of ecological systems theory, Incorporation of family-related risk factors. (2) Typicality: The selected family risk factors should be consistent with the actual situation of most families in China. (3) Relevance: The cumulative risk of the selected family should be closely related to non-suicidal self-injury. Some factors with small effects or obvious differences were not included in the study, such as parents’ occupational prestige. (4) Uniqueness: Ensure that family risk factors are unique and avoid obvious overlap between two or more risk factors. For example, parent-child communication is included in parent-child attachment and is not as typical as the selected risk factors. In summary, this study selected 7 types of cumulative family risks. The survey status of each cumulative family risk is explained below.
Parent-child attachment. The Parent-child Attachment Questionnaires developed by Bao and Xu were used to investigate parent-child attachment [35]. It has 25 items and is divided into three dimensions: trust, communication, and alienation. A 5-point Likert scale was used to rate the scores, with higher scores representing a higher level of parent-child attachment. In this study, Cronbach’s alpha was 0.87.
Parental conflict. The Children’s Perception of Interparental Conflict Scale developed by Chi and Xin was used to investigate children’s perceptions of parental conflict [36]. It has 17 items and is divided into three dimensions: conflict frequency, conflict intensity, and conflict resolution. A 4-point Likert scale was used to rate the scores, with higher scores representing a higher level of children’s perceptions of parental conflict. In this study, Cronbach’s alpha was 0.87.
Family financial pressure. The Family Financial Difficulties developed by Wang were used to investigate family financial pressure which contains 16 items [37]. A 5-point Likert scale was used to rate the scores, with higher scores representing a higher level of family financial pressure. In this study, Cronbach’s alpha was 0.87.
Parenting style. The short-Egna Minnen Barndoms Uppfostran Questionnaires, s-EMB developed by Jiang was used to investigate parenting style [38]. It has 21 items and is divided into three dimensions: rejection, emotional warmth, and overprotection. A 4-point Likert scale was used to rate the scores, with higher scores representing a higher level of effortful control. In this study, Cronbach’s alpha of rejection, emotional warmth, and overprotection were 0.88, 0.91, and 0.89 respectively.
Parental controls. The Parental Controls Questionnaires developed by Wang were used to investigate parental controls [39]. It has 18 items and is divided into three dimensions: causes feelings of guilt, withdrawal of love, and arbitrary power. A 5-point Likert scale was used to rate the scores, with higher scores representing a higher level of parental control. In this study, the Cronbach’s alpha was 0.88.
Parental emotional neglect. The Emotional Neglect Scale developed by Chen was used to investigate children’s perceptions of parental emotional neglect which contains 10 items [40]. A 4-point Likert scale was used to rate the scores, with higher scores representing a higher level of perceptions of parental emotional neglect. In this study, the Cronbach’s alpha was 0.88.
Parental maltreatment. The Parental Emaltreatment Scale developed by Zhao was used to investigate parental emaltreatment [41]. It has 23 items and is divided into four dimensions: emotional abuse, physical abuse, sexual abuse, and physical neglect. A 5-point Likert scale was used to rate the scores, with higher scores representing a higher level of perceptions of parental maltreatment. In this study, the Cronbach’s alpha was 0.88.