The Denver Youth Survey (DYS) is a longitudinal study that examines the causes, correlates, and consequences of delinquency, drug use, and other problem behaviors. The primary goal of the study is to identify social conditions, personal characteristics and developmental patterns linked to sustained involvement in delinquency and drug use over the life course.
The design of the DYS included the selection of respondents from socially disorganized neighborhoods of Denver, Colorado. Socially disorganized neighborhoods were identified by a formal social area analysis. In addition, respondents were selected only from neighborhoods with high crime rates.
All interviews were conducted in private settings, usually in the respondent's home. The vast majority of interviews were conducted in a face-to-face format, although for later waves, respondents who had moved some distance from the research site (nationally or internationally) were interviewed by telephone under strict privacy rules. All interviews (child, youth, young adult, and parent) were conducted after obtaining informed consent/assent from the respondent and from a parent of all respondents under the age of 18.
For each Wave, the DYS interviews include a Main Parent Interview that includes family information about the family (marital status of parents, parental employment, family income, etc.) and information about one child/youth respondent. If there was more than one participating child/youth in the household, then one supplemental interview was conducted for each remaining participating child/youth that asked for information about that specific child/youth.
For the purpose of analyses, the general family information applicable to all children/youth was integrated with the data from the supplemental interviews to create a set of complete data for each child/youth.
A total of 45,694 households were identified in the enumeration process. In the first sampling, 18,804 households were selected. The second or over-sampling of the "traditional disorganized" area included an additional sample of 1,473 from the 12,204 enumerated households in this stratum.
Sampled households were screened for the presence of eligible children and youth (those aged 7, 9, 11, 13, or 15 in 1987) by completing a household listing ("family chart") of all persons residing in a selected household and obtaining their birth dates. This screening information was obtained from an adult residing in the household. Households containing an eligible child/youth respondent were included in the sample, and interviews with all eligible children and youths, and one of their parents (principal caretaker) were attempted.
The screening process resulted in listings for 20,331 households or "listing lines" and in households in which two or more families were living, a separate family chart was obtained for each family. For the purpose of the study, each additional dwelling unit and each additional family in multiple family households resulted in an additional "listing line," so that all individual families that included an eligible child would each be interviewed as a separate family. Because of this, there are more households listed as being screened than the size of the original household sample. Also, during the screening effort, it was discovered that some structures identified as dwelling units during the enumeration process, were not in fact dwelling units. When such structures were selected as part of the household sample, they were considered to be "blank draws," to preserve the original sampling fraction and not affect the probability of selection of other households. The overall successful screening completion rate was approximately 92 percent.
Of 1,360 households that contained eligible youth, interviews were completed in 1,145 households. The remaining 215 households were "family refusals," where in most cases parents refused to participate for themselves and for their children. In a very few cases, the parents provided permission, but all of the children in the household/family refused to participate. Similar outcomes hold for families, which includes counts for families living in multiple family households. In 21 households, a child or youth interview was completed but an interview with the parent/caregiver for that household was not completed.
A total of 1,808 eligible youth resided in the successfully screened sampled households, of whom 1,528 completed an interview in the first year of the study, for a completion rate of 85 percent. A total of 1,132 parents/principal caretakers of these youth also completed 1,496 interviews about their participating children, for a completion rate among parents of participating youth of 98 percent.
Children and youth who were 7, 9, 11, 13, or 15 years old in 1987, and one of their parents, who lived in one of the more than 20,000 randomly selected households in Denver, Colorado.
The selection of variables was guided by the problem behaviors and by the rich mix of variables provided by a theoretical model guiding the project with a focus on biological history, neighborhood social disorganization and effect, conventional and deviant socialization and bonding, personality and mental health, peer influences, secondary (external) controls and rational choice. Topics covered in the data include:
- Measures of drug use and delinquency as well as other problem behavior and psychopathology
- Family demographics, neighborhood social characteristics and family integration and support within these neighborhoods
- Family variables including parenting, the child's involvement and attachment to the family, marital
discord, parent's domestic violence, and self-reported parental drug use and criminality, as well as
arrests of family members
- Child/youth involvement in school, community activities, religious activities, and work
- Personal and psychological characteristics of the child/youth
- Personal and psychological characteristics of parents
- Medical history including birth trauma, family medical and behavioral problems, developmental and
learning problems as well as current physical characteristics and nutrition
- Involvement with and the delinquent/drug use orientation of peers, including a special section on gangs
- Educational, occupational, and current social strain
- Secondary controls and rewards for drug use and delinquency, including risk of arrest and juvenile justice system
- Use of Drug Treatment and Mental Health Services by focal child/youth respondents and other family
- Measures for examining rational choice theory
The project obtained over a 90 percent retention rate in each of the first five years.