MICHIGAN YOUTH PROJECT

Project description and development:

Several factors determine the types of behavior and emotional problems children exhibit. These include sociocultural backgrounds of children's families and ensuing parenting practices. Parental attitudes and practices may discourage some types of child problems while fostering others. Child problem type and concomitant parental distress regarding these problems may determine whether clinical intervention is sought. Possible linkages between these factors were explored via surveys of clinic-referral problems of children of African descent in two different societies: (a) the African-British society of Jamaica which discourages child aggression and other externalizing behavior but encourages withdrawal and other internalizing behavior; and (b) the African-American society of Mid-Michigan which is of African-British heritage, but exists in the United States where aggression and other externalizing behavior are more widely accepted.

As a part of a large scale study funded through the David Walker Research Institute, Clinic records of African-American children and adolescents ages 4-18 in Mid-Michigan were surveyed from approximately 12 locations throughout the Detroit metropolitan area, as well as southern and western Michigan (i.e., Flint, Muskegan, Ann Arbor. Lansing, and Grand Rapids).   Sites included hospitals, and both in-patient and out-patient clinics that had African American child clients.  The data include demographic information, presenting problems, referral source, person reporting problems, diagnoses (if appropriate), and treatment recommendations.  Corresponding Jamaican data were collected from Kingston and other parts of Jamaica.

Parent-reported problems were recorded, coded and summed as internalizing, externalizing, and other problems according to an empirically derived classification system (i.e., Child Behavior Checklist; CBCL). A 2 (nationality) x 3 (age-group; 4-5, 6-11, 12-18) X 2 (gender) Analysis of Covariance with SES as a covariate was separately computed for problems summed under each category and for total number of problems reported. Following an earlier Jamaican clinic- survey procedure, problems of African-American children that do not fit the CBCL system are being recorded for the potential development of a classification system for future assessment and study of this African-American population.
 

Data files:

In SAS, use: sasuser.allmich6
Backed up in allmich6.bac; allmich4 and allmich5 are archive files
 

SAS program files which create datasets:

It is not possible to recreate the final dataset, because it was edited in sas;
However, these program files are useful for reference re: items, subscales, etc.

C:\data\old\michigan\clinic\programs\michpt2.sas

• reads in demo info, refer, etc, 15 problems reported, diagnoses, who gave,    treatment rec.
• creates diagnostic groups (e.g., diagnosis made by psychiatrist vs. psychologist)
• creates geographic groups (e.g., Detroit vs. Western Michigan)
• creates cbcl item scores from problem items
• creates cbcl subscales and total problem scores

• reads in ses data
• creates ses data subscales
• creates diagnostic categories
• creates treatment categories
• separates inpatient and outpatient

• reads in the 20 extra items which are not cbcl items (reported by more than 10 respondents)
• there is a poster on this on the left side of the second bookshelf (from bottom).

More information about this study can be found on the research page of the David Walker Research Institute web site.