Knowledge of Nonsuicidal Self-injury in Populations That Self-injure

Darcy Leanne Cates, Western Kentucky University

Abstract

Archived data were utilized for the present study which examined knowledge about non-suicidal self-injury, or NSSI, in individuals who engage in various degrees of the behavior and those who do not self-injure. Knowledge about NSSI was measured in three groups of respondents: those with no history of self-injurious behavior (no NSSI group), those with more limited experience with NSSI who reported 1-30 incidences of NSSI (limited NSSI group), and those with an extensive history (extensive NSSI group) who reported over 30 incidences of NSSI. To measure knowledge, participants were asked level of agreement with myths and facts about NSSI using Jeffery and Warm’s (2002) knowledge measure. It was hypothesized that the knowledge base would be higher in individuals with more extensive histories of NSSI. Further, individuals with limited histories of NSSI were predicted to have more knowledge than those who have never self-injured. Additionally, this study also hypothesized that the individual item response will vary; depending on extent of NSSI behavior.

Group mean scores on the measure were analyzed for differences using a one-way analysis of covariance (ANCOVA) while controlling for the differing group demographic variables of age, sexual orientation, and education level. Results indicated that individuals who have more extensive histories of NSSI evidenced higher mean scores on the measure when controlling for age, sexual orientation and educational level. Individuals with limited histories of NSSI evidenced lower mean scores, and those with no history of NSSI evidenced the lowest scores. In regard to individual item response, items were correlated with seven levels of NSSI (no NSSI, one incident of NSSI, 2-4 incidences, 5-10 incidences, 11-20 incidences, 21-30 incidences and more than 30 incidences). It was found that accuracy was significantly correlated with degree of self-injurious behaviors, with the exception of one item. This item and three additional items also produced weak correlations with other items on the measure. Each item is discussed with regard to group item performance and possible deletions in order to strengthen the measure.

Overall, the results of this investigation supported the reliability and validity of the Jeffery and Warm (2002) knowledge measure for use with individuals who self-injure. Results are discussed in relation to the need for accurate knowledge about NSSI, the importance of refining and strengthen the measure for this use, and additional research directions.