Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Purpose: Mental health professionals in schools include social workers, psychologists, counselors, and nurses. Assessments by psychologists, counselors, and nurses typically focus on specific domains of children's functioning—i.e., behavioral, academic, socio-emotional, and physical performance. Social work bridges the disciplines by emphasizing the impact of the social environment on all domains of functioning. With data on the social environment, school social workers can help school teams identify the most appropriate interventions to enhance student functioning in multiple areas. The current study identified five common profiles of social environmental risk and protection that may facilitate the development of appropriate combinations of individual- and group-level interventions in elementary schools.

Methods: Data were collected using the Elementary School Success Profile (ESSP). The analyzed sample comprised 532 3rd - 5th graders. Latent profile analysis with MPlus was used to identify profiles of social environmental risk and protection. Eight composite variables assessing child perceptions of the neighborhood (neighbors who care), school (teachers who care, school as a fun place to learn, school as a fun place to be with other children), friends (friends who care, accepted by peers, friends have good behavior), and family (family who care) were analyzed. Statistical and substantive criteria guided the choice of the best model. The final model was confirmed through the use of different start values and random sub-samples, and validated through the use of means comparisons on outcome variables.

Results: A five-class solution emerged as the best model, based on log likelihood and BIC values, class probabilities, class sizes, and substantive criteria. The final solution had a BIC of -4987.7, a loglikelihood value of 2657.0, entropy of .85, and average class probabilities ranging from .86 to .96. The five classes represented groups of students with different social environmental profiles requiring diverse prevention and intervention approaches. The first class (49.6% of the sample) had high (positive) scores on all 8 dimensions of the social environment. A second class (5.6% of the sample) had low to fair scores (55% to 74%) on all dimensions. A third class (7.1% of the sample) had mostly fair and moderate scores except for a low score (59%) in the family domain. The profile of the fourth class (31% of the sample) suggested risk only in the academic aspects of school. The profile of the final class (6.6% of the sample) indicated strengths in all domains except the friend domain, where peer rejection and antisocial behavior of friends were risk factors. Validating the class solution, means comparisons indicated consistently different levels of physical, psychological, social, and academic functioning across classes in expected directions.

Implications: The current analysis supported the relationship of the social environment to the outcomes of primary interest to different school-based professionals. It also suggested that there may be common profiles of social environmental risk and protection that can be used to organize relevant prevention and intervention efforts for subsets of elementary school children.