Abstract
Objective: The Maryland Assessment of Recovery in Serious Mental Illness (MARS) is a 25-item, 5-point Likert recovery scale that has been validated for use with adult clients. The present study examined the psychometrics of the MARS in an adolescent residential sample.
Methods: Adolescents (n = 121, 60% male, 52% Caucasian, 42% African American) were assessed periodically during residential treatment using MARS, K-SADS-PL diagnostic interviews, Youth Self Report (YSR), and Personal Well-Being Index (PWI-SC).
Results: T tests (MARS x gender) and one-way ANOVA (MARS x age, race, primary diagnosis) showed no differences in MARS scores for normative data. Reliability was strong using test retest (r = .914) and internal consistency (Chronbach’s a = .952). Repeated measures ANOVAs found significant main effects and interactions; as number of diagnoses declined from admission (M = 3.13) to discharge (M = 1.70), MARS scores significantly increased from admission (M = 93.97) to discharge (M = 105.90); as symptom severity (YSR total Problems) decreased from admission (M = 66.55) to discharge (M = 60.55), MARS scores increased showing sensitivity to change validity. Construct validity revealed a moderate correlation with PWI-SC (r =.42).
Conclusion: The present study provides preliminary evidence that the MARS is a psychometrically sound instrument for use in assessing recovery in adolescents. Several limitations of the study are discussed in the interest of further research on use of the MARS with adolescent clients.
Keywords: Personal recovery, assessment, adolescents, psychometrics, Maryland Assessment of Recovery in Serious Mental Illness (MARS).