Disconnected Youth: Does Operationalization Matter?

Background:

The transition to adulthood is a critical developmental period with implications for health and well-being throughout the life course.  In recent years, U.S. researchers and policy makers have identified a group of young people that may need additional supports in this period – those that are disconnected from education or the labor market. But, the way in which the “disconnected youth” population has been operationalized varies across studies resulting in different estimates of the size of the population and potentially altering its composition in significant ways. This study used nationally representative data to examine how three different measures used in prior research shape size estimates and characteristics of the disconnected youth population.

Methods: The sample was young adults (18-25) surveyed in the 2013 National Survey of Drug Use and Health (n=18,142). Three measures of disconnected youth were created. All three identified youth that were not working and not enrolled in school.  But, some looked at current employment versus current and past year and some excluded possible reasons for disconnection including marital/parenting status and educational attainment.  Category 1 excluded individuals with more than a high school education and used current employment status (n=2,159). Category 2 excluded married youth who were parenting and included those who were not in school and had not worked in the past year (n=1,164).  Category 3 included all those who were not in school and had not worked in the past year, regardless of parenting status (n=1,401). Weighted frequencies provided estimates and characteristics of the population. Three separate logistic regressions examined predictors of each disconnection variable such as health, insurance status, drug and alcohol abuse or dependence, government assistance, and county size.

Results: Estimates of the percent of disconnected youth ranged from 6% (Category 2) to 11% (Category 1) and characteristics of those belonging in the three categories varied. Females were overrepresented in all categories, particularly in Category 3, and all categories included disproportionate numbers of black youth. Living in a non-metro county was positively associated with being in all three categories, as was being poor, receiving government assistance, and being arrested. Reporting excellent health and drug or alcohol abuse or dependence were negatively associated with all three disconnected categories. Gender was only a significant predictor for Category 3.

Conclusions/Implications: Differences in the size and composition of the disconnected youth population based on how “disconnected” is operationalized highlight important considerations. Broader measurements capture a larger group of young people, but there were surprisingly few differences in the characteristics among the three groups.  Regardless of operationalization, females were overrepresented. Females, especially mothers, may be a subpopulation of disconnected youth in need of enhanced attention. Additionally, looking at current employment (i.e., within the last week) resulted in slightly different predictors and differences in the magnitude of association with disconnectedness. These findings suggest that operationalization of ‘disconnected youth’ produces widely varying estimates on the number of disconnected youth in need of intervention but that the characteristics of these groups are perhaps largely the same.