The Massachusetts Youth Screening Instrument-Version 2 (MAYSI-2) is a brief, self-report screening tool designed to identify youths with special mental health needs in the juvenile justice system. It is not a diagnostic tool. Instead, it identifies emergent risks of youths in need of a comprehensive psychological assessment. The developers of the MAYSI-2 designed it for administration by nonclinicians and normed it for use with youths aged 12 to 17 years on entry into one of three different settings in the juvenile justice system—intake probation, pretrial detention, and postsentencing correctional facilities. It is one of the only true mental health “screening” tools (as opposed to assessment tools) developed for juvenile justice settings. As such, the MAYSI-2 is important for juvenile justice administrators who have an obligation to manage the serious mental health needs of youths in their care by, among other procedures, implementing mental health screening. At the time of writing this entry, the MAYSI-2 was being used statewide by juvenile justice agencies in more than 35 states.
MAYSI-2 Description and Administration
The MAYSI-2 is a 52-item, “yes/no” screening tool on which youths report the presence or absence of symptoms or behaviors related to several areas of emotional, behavioral, and psychological disturbances experienced “within the past few months.” The test can be completed in 10 to 15 minutes and is generally administered by frontline, nonclinical staff via a voice computer program (MAYSIWARE) or a paper-and-pencil test, both of which are available in English and Spanish. The MAYSI-2 is self-administered (youths read the questions and circle their answers) unless youths have reading difficulties and the voice com-puter program is not an option. In these cases, staff should read questions to the youths and allow them to circle their own answers. The pencil-and-paper and voice computer modes of administration appear to yield comparable MAYSI-2 scores.
The 52 items produce scores on six clinical scales— Alcohol/Drug Use (ADU), Angry/Irritable (AI), Depressed/Anxious (DA), Somatic Complaints (SC), Suicide Ideation (SI), and Thought Disturbance (TD; for boys only)—and one nonclinical scale—Traumatic Experiences (TE), which screens for reported exposure to potentially traumatic events. Scale scores are generated from simple sums of the items, which range from 5 to 9 items depending on the scale. The TD scale is for boys only because factor analyses could not derive a coherent TD scale for girls, and the item content of the TE scale differs for boys and girls.
Each of the six clinical scales has a “Caution” cutoff to signal a “clinically significant” elevation and a “Warning” cutoff, which was based on the scores separating the upper 10% of youths in the development sample. The constellation of Caution and Warning cutoffs, which should signal a response for a particular youth (i.e., decision-making rules), was not prescribed by the MAYSI-2 developers. Instead, these decisions are left to the discretion of the juvenile justice site based on their resources and needs to respond to youths in their care.
MAYSI-2 Development and Factor Structure
The MAYSI-2 was created by Thomas Grisso and Richard Barnum, who selected a pool of items related to mental disorders, emotional disturbances, and behavioral problems common to adolescents. The final 52 items were generated from pilot testing on a small sample of youths. The developers identified the seven MAYSI-2 scales from factor analyses performed on data from 1,279 juvenile-justice-involved youths in Massachusetts. They derived Caution cutoff scores based on the optimal balance between sensitivity and specificity in identifying youths in this sample with clinical elevations on the Millon Adolescent Clinical Inventory and the Youth Self-Report. Researchers cross-validated the MAYSI-2 factor structure and indices of internal consistency using a sample of more than 4,000 juveniles from California. Recent studies reported almost identical factor structures using confirmatory factor analytic techniques.
MAYSI-2 Reliability and Validity
MAYSI-2 scales have acceptable internal consistency, with Cronbach’s alpha coefficients ranging from .61 (TD scale) to .86 (ADU scale). Alpha coefficients are comparable between genders and racial groups (with a few exceptions) and have been replicated across studies. Test-retest reliability estimates (intraclass correlation coefficients), based on comparisons of MAYSI-2 scores soon after admission to a detention facility to scores 5 to 8 days, later range from .53 (SC scale) to .89 (AI scale).
The national norm study for the MAYSI-2 found that girls were 1.5 to 2.3 times as likely as boys to score above the Caution on every scale except the ADU scale. This finding was consistent across more than 200 juvenile justice settings from around the United States regardless of youths’ age, race, or legal status. Also consistent was that Whites had higher odds of meeting Caution on the SI, SC, and ADU scales than Blacks or Hispanics.
Concurrent validity studies show that MAYSI-2 scales correlate with other adolescent mental health scales in the expected direction. However, most MAYSI-2 scales do not map directly onto diagnostic scales of other measures. This is likely because MAYSI-2 scales are primarily heterotypic and measure symptoms that would span multiple diagnoses (e.g., anger). Predictive validity studies indicate that MAYSI-2 scores predict several institutional outcomes such as institutional violence, lengthier sentences, staff interventions, and service provision. Pre-post studies indicate that adoption of the MAYSI-2 in detention facilities can significantly decrease violent incidents, suicide attempts, and other areas of maladjustment.
Some issues could benefit from further research. First, it is unknown how long MAYSI-2 scores can be considered valid. MAYSI-2 scores were not intended to have long-term stability given the items measure acute symptoms, which are expected to fluctuate. Second, it is unclear how the timing of MAYSI-2 administration may affect scores. Evidence suggests that youths receiving it within the first few hours of admission have lower scores than those taking it a day or two later. Finally, the developers should report on the psychometric properties of the Spanish-language version as data become available.
- Grisso, T., & Barnum, R. (2006). Massachusetts Youth Screening Instrument—”Version 2: User’s manual and technical report. Sarasota, FL: Professional Resource Press.
- Grisso, T., Barnum, R., Fletcher, K. E., Cauffman, E., & Peuschold, D. (2001). Massachusetts Youth Screening Instrument for mental health needs of juvenile justice youths. Journal of the American Academy of Child & Adolescent Psychiatry, 40(5), 541-548.
- Grisso, T., & Quinlan, J. C. (2005). Massachusetts Youth Screening Instrument—Version 2. In T. Grisso, G. Vincent, & D. Seagrave (Eds.), Mental health screening and assessment in juvenile justice (pp. 99-111). New York: Guilford Press.