U.S. flag

An official website of the United States government, Department of Justice.

Psychometric Properties of the Mental Health Screening Form III Within a Metropolitan Jail

NCJ Number
Criminal Justice and Behavior Volume: 36 Issue: 6 Dated: June 2009 Pages: 607-619
Mark A. Ruiz; Roger H. Peters; Gretchen M. Sanchez; Jan P. Bates
Date Published
June 2009
13 pages
This study assessed the psychometric performance and convergent validity of a screening instrument for mental health disorders, the Mental Health Screening Form III (MHSF-III), using drug-involved inmates in a large metropolitan jail as a test sample (n=332).
A high degree of internal consistency showed the instrument's reliability; this is consistent with previously reported test-retest findings (Carroll and McGinley, 2001; Sacks et al., 2007). The items also had acceptable mean inter-item correlations, indicating that despite being correlated, they were not overly redundant. The lack of statistical redundancy in the items is consistent with their diverse content. The majority of the MHSF-III items displayed good item-total correlations and endorsement frequencies. Item-total correlation, an indicator of the extent to which the item is linked with the total score of the instrument, shows that many of the items have a good association with mental illness. Most of the items also had moderate endorsement frequencies. Items that are endorsed by most respondents (i.e., high endorsement frequencies) are not optimal for mental health screening and can increase inappropriate referrals. Conversely, items that are rarely endorsed may increase false-negative screening results. Most of the MHSF-III items performed well for screening purposes. This assessment provided additional evidence of the convergent validity of the MHSF-III. The total score correlated with self-reported history of trauma. An advantage of the MHSF-III is that it covers a wider range of symptoms than instruments such as the Brief Jail Mental Health Screen (BJMHS). Many of the symptoms identified by the MHSF-III, such as suicidal or aggressive behavior, are critical for managing offenders. Ancillary analyses indicate that items with questionable functioning could be dropped with little to no effect on the reliability and validity of the instrument. Future work should examine whether the instrument's reliability and validity generalizes across various populations (e.g., racial and ethnic categories). 2 tables and 57 references