COSS™

Assessment Reports provide detailed information about scores from a single administration, presented both numerically and graphically. An individual's scores are compared to those in the normative sample and elevations at the scale and subscale level are indicated.
Progress Reports compare the results of two to four administrations for the same individual to measure changes over time. These reports are ideal to use when monitoring treatment and intervention effectiveness.
Comparative Reports combine the results of different raters to provide an overview of an individual's scores from a multi-rater perspective. This highlights potentially important inter-rater differences in scores.

Rating scales are available in parent, teacher, and self-report form formats. The COSS scales include Task Planning, Organizes Actions, and Memory and Materials Management. An Inconsistency Index is available on all three forms. This index is useful for determining response patterns that can affect results due to careless/random responding.

The self-report form also contains a Positive Impression Index to help identify inaccurate representation of problems by making situations appear better than they really are. has a negative attitude toward the testing process.

Both test-retest reliability and internal consistency are very good for the COSS. Internal consistency coefficients for the total sample range from .70 to .98, and 2- to 4-week test-retest reliability (Cronbach’s alpha) range from .88 to .99 (all correlations significant, p < .001). Support for the validity of the structure of the COSS forms was obtained using factor analytic techniques on derivation and confirmatory samples. Convergent and divergent validity were supported by examining the relationship between COSS scores and the Conners 3rd Edition^™ (Conners 3^™). Discriminative validity analyses were also supported by examining how accurately COSS scores discriminate between children from relevant clinical groups (e.g., ADHD).

Over 5,000 assessments were collected, including ratings of youth in the general population sample, clinical cases, and those included in the validity studies. The normative sample for the COSS includes ratings from 480 parents, 480 teachers, and 480 youth. The sample was stratified to represent the general U.S. population according to age, gender, and race/ ethnicity (U.S. Bureau of the Census, 2000). The normative sample is also diverse in terms of parental education level and geographic region. Separate norms are available for males and females, in 2-year age intervals (combined gender norms are also available).