Four studies evaluate the new Self Description Questionnaire II short-form (SDQII-S) that measures 11 dimensions of adolescent self-concept based on responses to 51 of the original 102 SDQII items and demonstrate new statistical strategies to operationalize guidelines for short-form evaluation proposed by G. T. Smith, D. M. McCarthy, and K. G. Anderson (2000). Multiple-group confirmatory factor analyses revealed that the factor structure based on responses to 51 items by a new cross-validation group (n=9,134) was invariant with the factor structures based on responses to the same 51 items and to all 102 items by the original normative archive group (n = 9,187). Reliabilities for the 11 SDQII-S factors were nearly the same and consistently high (.80 to .89) for both groups. Multitrait-multimethod analyses support the internal validity of responses over time. Gender and age effects on the 11 SDQII-S factors were invariant across the archive and cross-validation groups.

Existing research posits multiple dimensions of bullying and victimization but has not identified well-differentiated facets of these constructs that meet standards of good measurement: goodness of fit, measurement invariance, lack of differential item functioning, and well-differentiated factors that are not so highly correlated as to detract from their discriminant validity and substantive usefulness in school settings. Here we demonstrate exploratory structural equation modeling, an integration of confirmatory factor analysis and exploratory factor analysis. On the basis of responses to the 6-factor Adolescent Peer Relations Instrument (verbal, social, physical facets of bullying and victimization), we tested invariance of factor loadings, factor variances-covariances, item uniquenesses, item intercepts (a lack of differential item functioning), and latent means across gender, year in school, and time. Using a combination of relations with student characteristics and a multitrait-multimethod analysis, we showed that the 6 bully/victim factors have discriminant validity over time and in relation to gender, year in school, and relevant psychosocial correlates (e.g., depression, 11 components of academic and nonacademic selfconcept, locus of control, attitudes toward bullies and victims). However, bullies and victims are similar in many ways, and longitudinal panel models of the positive correlations between bully and victim factors suggest reciprocal effects such that each is a cause and an effect of the other.The present investigation is a substantive-methodological synergy (Marsh & Hau, 2007). It brings to bear new, strong, and evolving methodology in order to evaluate the psychometric properties of a new multidimensional instrument designed to measure bullying and victimization and to tackle complex substantive issues in this area of research. Although bullying and victimization are widely claimed to be multidimensional constructs, previous attempts to measure a priori factors based on a strong measure have met with only limited success. Here, as in other areas of research in psychology, theory, good measurement, research, and practice are inexorably related such that the neglect of one will undermine pursuit of the others. A particularly important starting point is the development of a solid measure of bullying and victimization, based on sound theory and supported by empirical research demonstrating support for convergent and discriminant validity. In this article, we begin with a review of substantive

Relations between self-concept and mental health are best understood from a multidimensional perspective. For responses by 903 adolescents (mean age = 12.6) to a new French translation of the Self Description Questionnaire II (SDQII), confirmatory factor analysis demonstrated a well-defined multidimensional factor structure of reliable, highly differentiated self-concept factors. Correlations between 11 SDQII factors and 7 mental health problems (Youth Self-Report; YSR) varied substantially (.11 to -.83; mean r = -.35). Single higher-order factors could not explain relations among SDQII factors, among YSR factors, or between the SDQII and YSR factors. This highly differentiated multivariate pattern of relations supports a multidimensional perspective of self-concept, not the unidimensional perspective still prevalent in mental health research and assessment.

Aggressive Troublemaker (getting into physical fights, getting into trouble, being seen as a troublemaker, and being punished for getting into trouble) and Victim (being threatened with harm, not feeling safe) factors were related to 3 components of self-concept (General, Same Sex, and Opposite Sex) based on the large, nationally representative National Education Longitudinal Study of 1988 database. At 8th-, 10th-, and 12th-grade levels, Troublemaker and Victim constructs were reasonably stable over time and moderately positively correlated (many students were both troublemakers and victims). The Victim factor was negatively correlated with self-concept and had negative effects on subsequent self-concept. Whereas the Troublemaker factor was also correlated somewhat negatively with self-concept, it had small positive effects on subsequent self-concept: Low self-concept may trigger troublemaking behavior in a possibly successful attempt to enhance subsequent self-concept. Although boys had higher Troublemaker and Victim scores than did girls, the effects of these constructs on subsequent self-concepts were similar for boys and girls.School aggression and acting out behaviors are a significant problem for schools around the world. School aggressive troublemaking (e.g., not following school rules, getting into physical fights, and being perceived as a troublemaker) may increase the likelihood of sanctions, risk of school failure (Wells & Rankin, 1983), and predispositions for criminal activity as adults (Babinski, Hartsought, & Lambert, 1999). Being a victim of school troublemakers and violence is associated with underachievement, psychological distress, depression, psychopathology, internalizing behaviors, and deteriorating physical health (Hawker & Boulton, 2000;Olweus, 1997).The causes of and solutions to school troublemaking and victimization have been under intense scrutiny over the past 20 years (Smith & Brain, 2000). Gender differences have been found. Boys are more likely than girls to be troublemakers as well as victims (Boulton & Smith, 1994;Rigby & Slee, 1993). Girls, however, tend to engage in more covert forms of indirect aggressive behaviors, such as rumors, social rejection, and exclusion (Crick, Bigbee, & Howes, 1996). Aggressive troublemaking behaviors are stable over time (Kumpulainen, Rasanen, & Henttonen, 1999;Salmivalli, Lappalainen, & Lagerspetz, 1998), although they are more frequent among younger than older students, and the same person can be both a perpetrator and a victim (Bosworth, Espelage,

No abstract